Upload tokenizer and modeling_openmoe.py

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,3 +1,123 @@
 ---
 license: apache-2.0
 ---
+<p align="center">
+<img width="150px" alt="OpenMoE" src="https://github.com/XueFuzhao/OpenMoE/blob/main/logo.jpg?raw=true">
+</p>
+<p align="center"><a href="https://github.com/XueFuzhao/OpenMoE/tree/main">[Github]</a>  |  <a href="https://colab.research.google.com/drive/1xIfIVafnlCP2XVICmRwkUFK3cwTJYjCY#scrollTo=62T-2mH_tsjG">[Colab Demo]</a> | <a href="https://huggingface.co/OrionZheng">[Huggingface]</a> | <a href="https://discord.gg/bjGnGfjegU">[Discord]</a>  |  <a href="https://twitter.com/xuefz/status/1693696988611739947?s=61&t=Xc2k2W7vU_hlpNizGDCmOw">[Twitter]</a> | <a href="https://xuefuzhao.notion.site/Aug-2023-OpenMoE-v0-2-Release-43808efc0f5845caa788f2db52021879">[Blog]</a></p>
+</p>
+<hr>
+
+# OpenMoE-8B(890B tokens)
+OpenMoE is a project aimed at igniting the open-source MoE community! We are releasing a family of open-sourced Mixture-of-Experts (MoE) Large Language Models.
+
+Our project began in the summer of 2023. On August 22, 2023, we released the first batch of intermediate checkpoints (OpenMoE-base&8B), along with the data and code [[Twitter]](https://twitter.com/xuefz/status/1693696988611739947?s=61&t=Xc2k2W7vU_hlpNizGDCmOw). Subsequently, the OpenMoE-8B training was completed in November, 2023. After that, we embarked on explorations on 34B scale model, which is still ongoing.
+
+As a small student team, instead of pursuing the best model with better data, computation, and human power, we devote to fully sharing our training data, strategies, model architecture, weights, and everything we have with the community. We hope this project will promote research on this promising field and invite more contributors to work on open-sourced MoE projects together!
+
+[2024.01.12] The paper for the project and more evaluations are underway. For more information about the model, training, and evaluations, please visit our GitHub [repository](https://github.com/XueFuzhao/OpenMoE/tree/main).
+
+
+## Model Weights
+Currently, three models are released in total: OpenMoE-base, OpenMoE-8B(and its chat version), and OpenMoE-34B(intermediate checkpoint at 200B tokens). 
+
+| Model Name     | Description                      | #Param   |Huggingface |
+|----------------|-------------------------------------------------|----------|-------------|
+| **OpenMoE-8B(1.1T)**   | 8B MoE with comparable FLOPs of a 1.6B LLaMA(No SFT)  |8B        |[Link](https://huggingface.co/OrionZheng/openmoe-8b) |
+| **OpenMoE-8B-Chat (1.1T+SFT)**   | OpenMoE-8B-1.1T supervised finetuned on the [WildChat GPT-4 Subset](https://huggingface.co/datasets/allenai/WildChat-nontoxic)   |8B        |[Link](https://huggingface.co/OrionZheng/openmoe-8b-chat) |
+| **OpenMoE-34B (200B)**   |  34B MoE with comparable FLOPs of a 7B LLaMA(No SFT)  |34B        |[Link](https://huggingface.co/OrionZheng/openmoe-34b-200B) |
+
+Besides, we also release all our intermediate checkpoints for research purposes:  
+
+| Model Name     | Description                      | #Param   |Huggingface |
+|----------------|-------------------------------------------------|----------|-------------|
+| OpenMoE-8B-200B   | 8B MoE with comparable FLOPs of a 1.6B LLaMA(No SFT) |8B        |[Link](https://huggingface.co/OrionZheng/openmoe-8b-200B) |
+| OpenMoE-8B-400B   | 8B MoE with comparable FLOPs of a 1.6B LLaMA(No SFT)  |8B        |[Link](https://huggingface.co/OrionZheng/openmoe-8b-400B) | 
+| OpenMoE-8B-600B   | 8B MoE with comparable FLOPs of a 1.6B LLaMA(No SFT) |8B        |[Link](https://huggingface.co/OrionZheng/openmoe-8b-600B) |
+| OpenMoE-8B-800B   | 8B MoE with comparable FLOPs of a 1.6B LLaMA(No SFT)  |8B        |[Link](https://huggingface.co/OrionZheng/openmoe-8b-800B) | 
+| OpenMoE-8B-1T   | 8B MoE with comparable FLOPs of a 1.6B LLaMA(No SFT)  |8B        |[Link](https://huggingface.co/OrionZheng/openmoe-8b-1T) | 
+| OpenMoE-base   | A small MoE model for debugging only       |637M      |[Link](https://huggingface.co/OrionZheng/openmoe-base) |  
+| OpenLLaMA-base | A dense counter-part of OpenMoE-base            |310M      |[Link](https://huggingface.co/fuzhao/OpenLLaMA_Base) |
+
+The base model, which were trained using 128 billion tokens, served primarily for debugging purposes. After validating the effectiveness of our model architecture, we did not pursue further training. Consequently, their performance might not be very well, and the checkpoint are not suitable for practical applications. Better performence can be oberved from our 8B or 34B versions.
+
+The OpenMoE-8B with 4 MoE layers and 32 experts has been trained by 1.1T tokens. The SFT version has also been released after we finetuned the OpenMoE-8B-1.1T on the [wildchat]((https://huggingface.co/datasets/allenai/WildChat-nontoxic)) dataset's GPT-4 subset. Besides, we also provide some intermediate checkpoints at 200B and 890B tokens for research purposes.
+
+We are still training our OpenMoE-34B, which is a MoE model with 8 MoE layer and 32 experts. We released the intermediate checkpoint trained on 200B tokens on huggingface. If you are interested in the latest checkpoint, please feel free to drop Fuzhao an email ([email protected]).
+
+## Get Started
+
+### Inference with Pytorch
+Our PyToch implementation is supported by [Colossal AI](https://github.com/hpcaitech/ColossalAI). You can install our forked version directly for easier setup:
+```
+# Python version: 3.10.12
+# Install ColossalAI
+git clone --branch my_openmoe https://github.com/Orion-Zheng/ColossalAI.git
+pip install ./ColossalAI
+python -m pip install -r ./ColossalAI/examples/language/openmoe/requirements.txt
+```
+
+Then, you can inference by the following code on a A100 80GB machine.
+```
+from transformers import AutoTokenizer, AutoConfig, AutoModelForCausalLM
+
+model_path = "ckpts/openmoe-8b-chat"
+config = AutoConfig.from_pretrained(model_path)
+tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained(
+    model_path,
+    torch_dtype=torch.bfloat16,
+    trust_remote_code=True, 
+    device_map='auto'
+    )
+query = 'Question: How do I kill a process? Answer:'
+prompt = f'''<<SYS>>
+You are a helpful, respectful and honest assistant.
+<</SYS>>
+
+<s>[INST] {query} [/INST]'''
+
+inputs = tokenizer(prompt, return_tensors="pt").to('cuda')
+sample = model.generate(**inputs, max_new_tokens=32)
+print(tokenizer.decode(sample[0]))
+```
+
+
+If you don't have GPUs on your hand, don't worry! you can still experience our model on Colab(Note: this require a $10 Colab Pro Plan). You can experiment with OpenMoE-8B-Chat on Colab directly by [this](https://colab.research.google.com/drive/1xIfIVafnlCP2XVICmRwkUFK3cwTJYjCY). 
+- Running OpenMoE-8B requires ~49GB of memory in float32 or ~23GB in bfloat16. It can be executed on a Colab `CPU High-RAM`(in float32) runtime or an `A100-40GB`(in bfloat16) runtime, both of which require Colab Pro. The float16 precision is not recommended because sometimes it will lead to performance degradation.
+- Runing the OpenMoE-34B requries ~89GB of memory in bfloat16 or ~180GB in float32. To perform inference on multiple devices/offloading model weights to RAM, please refer to the script [here](https://github.com/XueFuzhao/OpenMoE/blob/main/script/inference_on_multi_devices.py).
+- A more detailed env setup script can be found [here](https://github.com/XueFuzhao/OpenMoE/blob/main/env/prepare_env.sh), or if you use docker, you can refer to the dockerfile [here](https://github.com/XueFuzhao/OpenMoE/blob/main/env/openmoe_infer_dockerfile). Note: you don't need t5x and Jax dependency if you are using our [huggingface ckpts](https://huggingface.co/OrionZheng/openmoe-8b-chat) without converting the jax checkpoints.
+
+Besides, we also provide a Colab [tutorial](https://colab.research.google.com/drive/1eIT1rtG7pORRQAYtQoMOAekUg7aZLDdn) demonstrating the jax checkpoint conversion. 
+
+
+## License
+
+Our code is under Apache 2.0 License.
+
+Since the models are trained on The Redpajama and The Stack dataset, please check the license of these two datasets for your model usage.
+
+
+## Authors
+
+This project is currently contributed by the following authors:
+
+[Fuzhao Xue](https://xuefuzhao.github.io/), [Zian Zheng](https://zheng-zian-andy.com), [Yao Fu](https://franxyao.github.io/), [Jinjie Ni](http://jinjie.one/), [Zangwei Zheng](https://zhengzangw.github.io/), [Wangchunshu Zhou](https://michaelzhouwang.github.io/), [Yang You](https://www.comp.nus.edu.sg/~youy/)
+
+## Acknowledgement
+The computational resources for this project were generously provided by the [Google TPU Research Cloud(TRC)](https://sites.research.google/trc/about/). We extend our heartfelt thanks to TRC for their invaluable support, which has been fundamental to the success of our work. Besides, we are extremely grateful to the [ColossalAI Team](https://github.com/hpcaitech/ColossalAI) for their tremendous support with the PyTorch implementation, especially [Xuanlei Zhao](https://oahzxl.github.io/) and [Wenhao Chen](https://github.com/CWHer), making training and inference of OpenMoE on GPUs a reality.
+
+## Citation
+
+Please cite the repo if you use the model and code in this repo.
+
+```bibtex
+@misc{openmoe2023,
+  author = {Fuzhao Xue, Zian Zheng, Yao Fu, Jinjie Ni, Zangwei Zheng, Wangchunshu Zhou and Yang You},
+  title = {OpenMoE: Open Mixture-of-Experts Language Models},
+  year = {2023},
+  publisher = {GitHub},
+  journal = {GitHub repository},
+  howpublished = {\url{https://github.com/XueFuzhao/OpenMoE}},
+}
+```

modeling_openmoe.py

@@ -0,0 +1,1140 @@
+# coding=utf-8
+# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" PyTorch OpenMoE model."""
+import math
+from typing import List, Optional, Tuple, Union
+
+import torch
+import torch.nn.functional as F
+import torch.utils.checkpoint
+from torch import nn
+from transformers.modeling_outputs import BaseModelOutputWithPast, CausalLMOutputWithPast
+from transformers.modeling_utils import PreTrainedModel
+from transformers.models.llama.configuration_llama import LlamaConfig
+# from .llama_attn import LlamaAttention
+
+from transformers.utils import (
+    add_start_docstrings,
+    add_start_docstrings_to_model_forward,
+    logging,
+    replace_return_docstrings,
+)
+
+from colossalai.kernel.cuda_native.mha.flash_attn_2 import HAS_FLASH_ATTN
+from colossalai.kernel.triton.llama_act_combine_kernel import HAS_TRITON
+from colossalai.moe.layers import SparseMLP
+from colossalai.moe.manager import MOE_MANAGER
+from colossalai.moe.utils import get_activation, set_moe_args
+
+
+
+if HAS_TRITON:
+    from colossalai.kernel.triton.llama_act_combine_kernel import LlamaActCombine
+
+logger = logging.get_logger(__name__)
+
+_CONFIG_FOR_DOC = "LlamaConfig"
+
+
+def set_openmoe_args(
+    config: LlamaConfig,
+    num_experts: int,
+    moe_layer_interval: int,
+    router_topk: int = 2,
+    router_capacity_factor_train: float = 1.25,
+    router_capacity_factor_eval: float = 2.0,
+    router_min_capacity: int = 4,
+    router_noisy_policy: str = None,
+    router_drop_tks: bool = True,
+    router_aux_loss_factor: float = 0.01,
+    router_z_loss_factor: float = 0.0001,
+    mlp_gated: bool = True,
+    label_smoothing: float = 0.001,
+    z_loss_factor: float = 0.01,
+    enable_load_balance: bool = False,
+    load_balance_tolerance: float = 0.1,
+    load_balance_beam_width: int = 8,
+    load_balance_group_swap_factor: float = 0.4,
+    enable_kernel: bool = False,
+    enable_comm_overlap: bool = False,
+    enable_hierarchical_alltoall: bool = False,
+) -> None:
+    """
+    MoE related arguments.
+    It inserts the MoE arguments into the Llama config.
+
+    Args:
+        config (LlamaConfig): Transformers Llama config.
+        num_experts (int, optional): Number of experts.
+        moe_layer_interval (int, optional): The interval moe layer.
+        router_topk (int, optional): Moe router top k. Defaults to 2.
+        router_capacity_factor_train (float, optional): Moe router max capacity for train. Defaults to 1.25.
+        router_capacity_factor_eval (float, optional): Moe router max capacity for eval. Defaults to 2.0.
+        router_min_capacity (int, optional): Moe router min capacity. Defaults to 4.
+        router_noisy_policy (str, optional): Moe router noisy policy. You can choose [Jitter, Gaussian, None]. Defaults to None.
+        router_drop_tks (bool, optional): Whether moe router drop tokens which exceed max capacity. Defaults to True.
+        router_aux_loss_factor (float, optional): Moe router aux loss. You can refer to STMoE for details. Defaults to 0.01.
+        router_z_loss_factor (float, optional): Moe router z loss. You can refer to STMoE for details. Defaults to 0.01.
+        mlp_gated (bool, optional): Use gate in mlp. Defaults to True.
+        label_smoothing (float, optional): Label smoothing. Defaults to 0.001.
+        z_loss_factor (float, optional): The final outputs' classification z loss factor. Defaults to 0.01.
+        enable_load_balance (bool, optional): Expert load balance. Defaults to False.
+        load_balance_tolerance (float, optional): Expert load balance search's difference tolerance. Defaults to 0.1.
+        load_balance_beam_width (int, optional): Expert load balance search's beam width. Defaults to 8.
+        load_balance_group_swap_factor (float, optional): Expert load balance group swap factor. Longer value encourages less swap. Defaults to 0.4.
+        enable_kernel (bool, optional): Use kernel optimization. Defaults to False.
+        enable_comm_overlap (bool, optional): Use communication overlap for MoE. Recommended to enable for muiti-node training. Defaults to False.
+        enable_hierarchical_alltoall (bool, optional): Use hierarchical alltoall for MoE. Defaults to False.
+    """
+    moe_args = dict(
+        num_experts=num_experts,
+        moe_layer_interval=moe_layer_interval,
+        router_topk=router_topk,
+        router_capacity_factor_train=router_capacity_factor_train,
+        router_capacity_factor_eval=router_capacity_factor_eval,
+        router_min_capacity=router_min_capacity,
+        router_noisy_policy=router_noisy_policy,
+        router_drop_tks=router_drop_tks,
+        router_aux_loss_factor=router_aux_loss_factor,
+        router_z_loss_factor=router_z_loss_factor,
+        mlp_gated=mlp_gated,
+        label_smoothing=label_smoothing,
+        z_loss_factor=z_loss_factor,
+        enable_load_balance=enable_load_balance,
+        load_balance_tolerance=load_balance_tolerance,
+        load_balance_beam_width=load_balance_beam_width,
+        load_balance_group_swap_factor=load_balance_group_swap_factor,
+        enable_kernel=enable_kernel,
+        enable_comm_overlap=enable_comm_overlap,
+        enable_hierarchical_alltoall=enable_hierarchical_alltoall,
+    )
+    set_moe_args(config, moe_args)
+
+
+# Copied from transformers.models.bart.modeling_bart._make_causal_mask
+def _make_causal_mask(
+    input_ids_shape: torch.Size, dtype: torch.dtype, device: torch.device, past_key_values_length: int = 0
+):
+    """
+    Make causal mask used for bi-directional self-attention.
+    """
+    bsz, tgt_len = input_ids_shape
+    mask = torch.full((tgt_len, tgt_len), torch.finfo(dtype).min, device=device)
+    mask_cond = torch.arange(mask.size(-1), device=device)
+    mask.masked_fill_(mask_cond < (mask_cond + 1).view(mask.size(-1), 1), 0)
+    mask = mask.to(dtype)
+
+    if past_key_values_length > 0:
+        mask = torch.cat([torch.zeros(tgt_len, past_key_values_length, dtype=dtype, device=device), mask], dim=-1)
+    return mask[None, None, :, :].expand(bsz, 1, tgt_len, tgt_len + past_key_values_length)
+
+
+# Copied from transformers.models.bart.modeling_bart._expand_mask
+def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Optional[int] = None):
+    """
+    Expands attention_mask from `[bsz, seq_len]` to `[bsz, 1, tgt_seq_len, src_seq_len]`.
+    """
+    bsz, src_len = mask.size()
+    tgt_len = tgt_len if tgt_len is not None else src_len
+
+    expanded_mask = mask[:, None, None, :].expand(bsz, 1, tgt_len, src_len).to(dtype)
+
+    inverted_mask = 1.0 - expanded_mask
+
+    return inverted_mask.masked_fill(inverted_mask.to(torch.bool), torch.finfo(dtype).min)
+
+
+def apply_rotary_embedding(q, k, cos, sin, decode=False, rotary_index=None):  
+    # q:  (bs, q_len, num_heads, head_dim)
+    # k:  (bs, q_len [+past_kv_len], num_heads, head_dim)
+    # cos: (max_seq_len, head_dim)
+    # sin: (max_seq_len, head_dim)
+    # rotary_index: (bs, 1)  # only used during decoding, when one query token is input at a time
+    """Helper function to apply Rotary Embeddings."""
+    cos = cos.to(q.dtype)
+    sin = sin.to(q.dtype)
+
+    if len(k.shape) == 3: # for multi query attention
+        k = k.unsqueeze(2)
+        multiquery = True
+    else:
+        multiquery = False
+
+    batch, qlen, qheads, d = q.shape
+    kbatch, klen, kheads, kd = k.shape
+    assert batch == kbatch, f"{batch} != {kbatch}"
+    assert d == kd, f"{d} != {kd}"
+    if decode and qlen == 1 and rotary_index is not None:
+        qcos = cos[rotary_index, :]  # (bs, 1, head_dim)
+        qsin = sin[rotary_index, :]  # (bs, 1, head_dim)
+        qcos = qcos.unsqueeze(2)  # (bs, q_len=1, 1, head_dim)  # broadcast to all heads
+        qsin = qsin.unsqueeze(2)  # (bs, q_len=1, 1, head_dim)    
+    else:
+        qcos, qsin = cos[:qlen, :], sin[:qlen, :]  # (q_len, head_dim)
+        qcos = qcos.unsqueeze(0).unsqueeze(2)  # (1, q_len, 1, head_dim)
+        qsin = qsin.unsqueeze(0).unsqueeze(2)
+    
+    kcos, ksin = cos[:klen, :], sin[:klen, :]  # (k_len, head_dim)
+    kcos = kcos.unsqueeze(0).unsqueeze(2)  # (1, k_len, 1, head_dim)  # broadcast to the whole batch, broadcast to all heads
+    ksin = ksin.unsqueeze(0).unsqueeze(2)  # (1, k_len, 1, head_dim)
+    out_q = (q * qcos) + (rotate_half(q) * qsin)
+    out_k = (k * kcos) + (rotate_half(k) * ksin)
+
+    if multiquery:
+        out_k = out_k.squeeze(2)
+
+    return out_q, out_k
+
+
+def rotate_half(x):
+    """Rotates half the hidden dims of the input."""
+    x1 = x[..., : x.shape[-1] // 2]
+    x2 = x[..., x.shape[-1] // 2 :]
+    return torch.cat((-x2, x1), dim=-1)
+
+class LlamaRMSNorm(nn.Module):
+    def __init__(self, hidden_size, eps=1e-6):
+        """
+        LlamaRMSNorm is equivalent to T5LayerNorm
+        """
+        super().__init__()
+        self.weight = nn.Parameter(torch.ones(hidden_size))
+        self.variance_epsilon = eps
+
+    def forward(self, hidden_states):
+        input_dtype = hidden_states.dtype
+        hidden_states = hidden_states.to(torch.float32)
+        variance = hidden_states.pow(2).mean(-1, keepdim=True)
+        hidden_states = hidden_states * torch.rsqrt(variance + self.variance_epsilon)
+        return self.weight * hidden_states.to(input_dtype)
+
+def SwiGLU(x):
+    """Gated linear unit activation function.
+    Args:
+        x : input array
+        axis: the axis along which the split should be computed (default: -1)
+    """
+    size = x.shape[-1]
+    assert size % 2 == 0, "axis size must be divisible by 2"
+    x1, x2 = torch.split(x, size // 2, -1)
+    return x1 * (x2 * torch.sigmoid(x2))
+
+
+class OpenMoeMLP(nn.Module):
+    def __init__(self, config: LlamaConfig):
+        super().__init__()
+        self.pretraining_tp = config.pretraining_tp
+        self.hidden_size = config.hidden_size
+        self.intermediate_size = config.intermediate_size
+        self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size * 2, bias=False)
+        self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False)
+        self.hidden_act = config.hidden_act
+        self.act_fn = get_activation(self.hidden_act)
+        self.use_kernel = config.enable_kernel
+
+    def forward(self, x):
+        if self.pretraining_tp > 1:
+            slice = self.intermediate_size // self.pretraining_tp
+            gate_proj_slices = self.gate_proj.weight.split(slice, dim=0)
+            up_proj_slices = self.up_proj.weight.split(slice, dim=0)
+            down_proj_slices = self.down_proj.weight.split(slice, dim=1)
+
+            gate_proj = torch.cat([F.linear(x, gate_proj_slices[i]) for i in range(self.pretraining_tp)], dim=-1)
+            up_proj = torch.cat([F.linear(x, up_proj_slices[i]) for i in range(self.pretraining_tp)], dim=-1)
+
+            intermediate_states = (self.act_fn(gate_proj) * up_proj).split(slice, dim=2)
+            down_proj = [F.linear(intermediate_states[i], down_proj_slices[i]) for i in range(self.pretraining_tp)]
+            down_proj = sum(down_proj)
+        else:
+            if HAS_TRITON and self.use_kernel and self.hidden_act == "swiglu":
+                down_proj = self.down_proj(LlamaActCombine.apply(self.gate_proj(x), self.up_proj(x)))
+            else:
+                down_proj = self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
+
+        return down_proj
+
+
+def repeat_kv(hidden_states: torch.Tensor, n_rep: int) -> torch.Tensor:
+    """
+    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
+    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
+    """
+    batch, num_key_value_heads, slen, head_dim = hidden_states.shape
+    if n_rep == 1:
+        return hidden_states
+    hidden_states = hidden_states[:, :, None, :, :].expand(batch, num_key_value_heads, n_rep, slen, head_dim)
+    return hidden_states.reshape(batch, num_key_value_heads * n_rep, slen, head_dim)
+
+
+class OpenMoeAttention(nn.Module):
+    """Multi-headed attention from 'Attention Is All You Need' paper"""
+
+    def __init__(self, config: LlamaConfig):
+        super().__init__()
+        self.config = config
+        self.hidden_size = config.hidden_size
+        self.num_heads = config.num_attention_heads
+        self.head_dim = config.head_dim
+        self.num_key_value_heads = config.num_key_value_heads
+        self.num_key_value_groups = self.num_heads // self.num_key_value_heads
+        self.pretraining_tp = config.pretraining_tp
+        self.max_position_embeddings = config.max_position_embeddings
+
+        self.q_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=False)
+        self.k_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=False)
+        self.v_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=False)
+        self.o_proj = nn.Linear(self.num_heads * self.head_dim, self.hidden_size, bias=False)
+        self.generate_fixed_pos_embedding(self.head_dim, self.max_position_embeddings, 1.0, 1e4)
+        self.use_kernel = config.enable_kernel
+        
+
+    def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
+        return tensor.view(bsz, seq_len, self.num_heads, self.head_dim).transpose(1, 2).contiguous()
+
+    def generate_fixed_pos_embedding(self, features, length, min_timescale=1.0, max_timescale=10000.0):
+        """Generate Sin/Cos for Rotary Embeddings.
+    
+        Args:
+          features: an integer
+          length: an integer
+          min_timescale: an optional float
+          max_timescale: an optional float
+    
+        Returns:
+          output_sin: a float32 Tensor with shape [length, features]
+          output_cos: a float32 Tensor with shape [length, features]
+        """
+        fraction = torch.arange(0, features, 2, dtype=torch.float32) / features
+        timescale = min_timescale * (max_timescale / min_timescale) ** fraction
+        rotational_frequency = 1.0 / timescale
+    
+        sinusoid_inp = torch.einsum("i,j->ij", torch.arange(length, dtype=torch.float32), rotational_frequency)
+    
+        sinusoid_inp = torch.cat([sinusoid_inp, sinusoid_inp], dim=-1)
+
+        self.register_buffer('sin', torch.sin(sinusoid_inp), persistent=False)  # persistent=False --> buffer won't appear in the state_dict
+        self.register_buffer('cos', torch.cos(sinusoid_inp), persistent=False)
+    
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Tuple[torch.Tensor]] = None,
+        output_attentions: bool = False,
+        use_cache: bool = False,
+    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
+        bsz, q_len, _ = hidden_states.size()
+
+        if self.pretraining_tp > 1:
+            key_value_slicing = (self.num_key_value_heads * self.head_dim) // self.pretraining_tp
+            query_slices = self.q_proj.weight.split((self.num_heads * self.head_dim) // self.pretraining_tp, dim=0)
+            key_slices = self.k_proj.weight.split(key_value_slicing, dim=0)
+            value_slices = self.v_proj.weight.split(key_value_slicing, dim=0)
+
+            query_states = [F.linear(hidden_states, query_slices[i]) for i in range(self.pretraining_tp)]
+            query_states = torch.cat(query_states, dim=-1)
+
+            key_states = [F.linear(hidden_states, key_slices[i]) for i in range(self.pretraining_tp)]
+            key_states = torch.cat(key_states, dim=-1)
+
+            value_states = [F.linear(hidden_states, value_slices[i]) for i in range(self.pretraining_tp)]
+            value_states = torch.cat(value_states, dim=-1)
+
+        else:
+            query_states = self.q_proj(hidden_states)
+            key_states = self.k_proj(hidden_states)
+            value_states = self.v_proj(hidden_states)
+
+        query_states = query_states.view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
+        key_states = key_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+        value_states = value_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+
+        kv_seq_len = key_states.shape[-2]
+        if past_key_value is not None:
+            kv_seq_len += past_key_value[0].shape[-2]
+        # cos, sin = self.rotary_emb(value_states, seq_len=kv_seq_len)
+        # query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin, position_ids)
+        if past_key_value is not None:
+            # reuse k, v, self_attention
+            key_states = torch.cat([past_key_value[0], key_states], dim=2)
+            value_states = torch.cat([past_key_value[1], value_states], dim=2)
+
+        past_key_value = (key_states, value_states) if use_cache else None
+
+        query_states = query_states.transpose(1, 2)
+        key_states = key_states.transpose(1, 2)
+        max_length = max(query_states.shape[1], key_states.shape[1])
+        assert max_length <= self.sin.shape[0]
+        sin, cos = self.sin[:max_length], self.cos[:max_length]
+        # TODO: for inference, we can add emb kv into cache to avoid computation
+        query_states, key_states = apply_rotary_embedding(
+            query_states, key_states, cos, sin, decode=True if q_len == 1 else False, rotary_index=position_ids
+        )
+        query_states = query_states.transpose(1, 2)
+        key_states = key_states.transpose(1, 2)
+
+        # repeat k/v heads if n_kv_heads < n_heads
+        key_states = repeat_kv(key_states, self.num_key_value_groups)
+        value_states = repeat_kv(value_states, self.num_key_value_groups)
+
+        if HAS_FLASH_ATTN and self.use_kernel:
+            from flash_attn import flash_attn_func  
+
+            query_states = query_states.transpose(1, 2)
+            key_states = key_states.transpose(1, 2)
+            value_states = value_states.transpose(1, 2)
+            attn_output = flash_attn_func(query_states, key_states, value_states, softmax_scale=1.0, causal=True)
+            attn_output = attn_output.transpose(1, 2).contiguous()
+        else:
+            attn_weights = torch.matmul(query_states, key_states.transpose(2, 3))
+
+            if attn_weights.size() != (bsz, self.num_heads, q_len, kv_seq_len):
+                raise ValueError(
+                    f"Attention weights should be of size {(bsz, self.num_heads, q_len, kv_seq_len)}, but is"
+                    f" {attn_weights.size()}"
+                )
+
+            if attention_mask is not None:
+                if attention_mask.size() != (bsz, 1, q_len, kv_seq_len):
+                    raise ValueError(
+                        f"Attention mask should be of size {(bsz, 1, q_len, kv_seq_len)}, but is {attention_mask.size()}"
+                    )
+                if self.training:
+                    attention_mask = attention_mask.clone().detach()
+                attention_mask[:, :, :, 0] = 0
+                attn_weights = attn_weights + attention_mask
+
+            # upcast attention to fp32
+            attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query_states.dtype)
+            attn_output = torch.matmul(attn_weights, value_states)
+
+        if attn_output.size() != (bsz, self.num_heads, q_len, self.head_dim):
+            raise ValueError(
+                f"`attn_output` should be of size {(bsz, self.num_heads, q_len, self.head_dim)}, but is"
+                f" {attn_output.size()}"
+            )
+
+        attn_output = attn_output.transpose(1, 2).contiguous()
+        attn_output = attn_output.reshape(bsz, q_len, self.num_heads * self.head_dim)
+
+        if self.pretraining_tp > 1:
+            attn_output = attn_output.split(self.hidden_size // self.pretraining_tp, dim=2)
+            o_proj_slices = self.o_proj.weight.split(self.hidden_size // self.pretraining_tp, dim=1)
+            attn_output = sum([F.linear(attn_output[i], o_proj_slices[i]) for i in range(self.pretraining_tp)])
+        else:
+            attn_output = self.o_proj(attn_output)
+
+        if not output_attentions:
+            attn_weights = None
+
+        return attn_output, attn_weights, past_key_value
+
+
+class OpenMoeDecoderLayer(nn.Module):
+    def __init__(self, config: LlamaConfig, moe: bool):
+        super().__init__()
+        self.hidden_size = config.hidden_size
+        self.moe = moe
+        self.self_attn = OpenMoeAttention(config=config)  
+#         self.self_attn = LlamaAttention(config=config)  # TODO: introduce LLaMA Positional Encoding
+        self.input_layernorm = LlamaRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        self.post_attention_layernorm = LlamaRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        if self.moe:
+            self.mlp = SparseMLP(
+                num_experts=config.num_experts,
+                hidden_size=config.hidden_size,
+                intermediate_size=config.intermediate_size,
+                router_top_k=config.router_topk,
+                router_capacity_factor_train=config.router_capacity_factor_train,
+                router_capacity_factor_eval=config.router_capacity_factor_eval,
+                router_min_capacity=config.router_min_capacity,
+                router_noisy_policy=config.router_noisy_policy,
+                router_drop_tks=config.router_drop_tks,
+                mlp_activation=config.hidden_act,
+                mlp_gated=config.mlp_gated,
+                enable_load_balance=config.enable_load_balance,
+                load_balance_tolerance=config.load_balance_tolerance,
+                load_balance_beam_width=config.load_balance_beam_width,
+                load_balance_group_swap_factor=config.load_balance_group_swap_factor,
+                enable_kernel=config.enable_kernel,
+                enable_comm_overlap=config.enable_comm_overlap,
+            )
+            self.pre_extra_mlp_layernorm = LlamaRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+            self.extra_mlp = OpenMoeMLP(config)
+        else:
+            self.mlp = OpenMoeMLP(config)
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Tuple[torch.Tensor]] = None,
+        output_attentions: Optional[bool] = False,
+        use_cache: Optional[bool] = False,
+    ) -> Tuple[torch.FloatTensor, Optional[Tuple[torch.FloatTensor, torch.FloatTensor]]]:
+        """
+        Args:
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
+            attention_mask (`torch.FloatTensor`, *optional*): attention mask of size
+                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
+            output_attentions (`bool`, *optional*):
+                Whether or not to return the attentions tensors of all attention layers. See `attentions` under
+                returned tensors for more detail.
+            use_cache (`bool`, *optional*):
+                If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding
+                (see `past_key_values`).
+            past_key_value (`Tuple(torch.FloatTensor)`, *optional*): cached past key and value projection states
+        """
+
+        residual = hidden_states
+
+        hidden_states = self.input_layernorm(hidden_states)
+
+        # Self Attention
+        hidden_states, self_attn_weights, present_key_value = self.self_attn(
+            hidden_states=hidden_states,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_value=past_key_value,
+            output_attentions=output_attentions,
+            use_cache=use_cache,
+        )
+        hidden_states = residual + hidden_states
+
+        # Fully Connected
+        residual = hidden_states
+        hidden_states = self.post_attention_layernorm(hidden_states)
+        hidden_states = self.mlp(hidden_states)
+        hidden_states = residual + hidden_states
+
+        if self.moe:
+            residual = hidden_states
+            hidden_states = self.pre_extra_mlp_layernorm(hidden_states)
+            hidden_states = self.extra_mlp(hidden_states)
+            hidden_states = residual + hidden_states
+
+        outputs = (hidden_states,)
+
+        if output_attentions:
+            outputs += (self_attn_weights,)
+
+        if use_cache:
+            outputs += (present_key_value,)
+
+        return outputs
+
+
+LLAMA_START_DOCSTRING = r"""
+    This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
+    library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
+    etc.)
+
+    This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
+    Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
+    and behavior.
+
+    Parameters:
+        config ([`LlamaConfig`]):
+            Model configuration class with all the parameters of the model. Initializing with a config file does not
+            load the weights associated with the model, only the configuration. Check out the
+            [`~PreTrainedModel.from_pretrained`] method to load the model weights.
+"""
+
+
+@add_start_docstrings(
+    "The bare LLaMA Model outputting raw hidden-states without any specific head on top.",
+    LLAMA_START_DOCSTRING,
+)
+class OpenMoePreTrainedModel(PreTrainedModel):
+    config_class = LlamaConfig
+    base_model_prefix = "model"
+    supports_gradient_checkpointing = True
+    _no_split_modules = ["OpenMoeDecoderLayer"]
+    _skip_keys_device_placement = "past_key_values"
+
+    def _init_weights(self, module):
+        std = self.config.initializer_range
+        if isinstance(module, nn.Linear):
+            module.weight.data.normal_(mean=0.0, std=std)
+            if module.bias is not None:
+                module.bias.data.zero_()
+        elif isinstance(module, nn.Embedding):
+            module.weight.data.normal_(mean=0.0, std=std)
+            if module.padding_idx is not None:
+                module.weight.data[module.padding_idx].zero_()
+
+    def _set_gradient_checkpointing(self, module, value=False):
+        if isinstance(module, OpenMoeModel):
+            module.gradient_checkpointing = value
+
+
+LLAMA_INPUTS_DOCSTRING = r"""
+    Args:
+        input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
+            Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
+            it.
+
+            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+            [`PreTrainedTokenizer.__call__`] for details.
+
+            [What are input IDs?](../glossary#input-ids)
+        attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
+
+            - 1 for tokens that are **not masked**,
+            - 0 for tokens that are **masked**.
+
+            [What are attention masks?](../glossary#attention-mask)
+
+            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+            [`PreTrainedTokenizer.__call__`] for details.
+
+            If `past_key_values` is used, optionally only the last `decoder_input_ids` have to be input (see
+            `past_key_values`).
+
+            If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`]
+            and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
+            information on the default strategy.
+
+            - 1 indicates the head is **not masked**,
+            - 0 indicates the head is **masked**.
+        position_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
+            config.n_positions - 1]`.
+
+            [What are position IDs?](../glossary#position-ids)
+        past_key_values (`tuple(tuple(torch.FloatTensor))`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`):
+            Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of shape
+            `(batch_size, num_heads, sequence_length, embed_size_per_head)`) and 2 additional tensors of shape
+            `(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)`.
+
+            Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention
+            blocks) that can be used (see `past_key_values` input) to speed up sequential decoding.
+
+            If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
+            don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
+            `decoder_input_ids` of shape `(batch_size, sequence_length)`.
+        inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
+            Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This
+            is useful if you want more control over how to convert `input_ids` indices into associated vectors than the
+            model's internal embedding lookup matrix.
+        use_cache (`bool`, *optional*):
+            If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
+            `past_key_values`).
+        output_attentions (`bool`, *optional*):
+            Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned
+            tensors for more detail.
+        output_hidden_states (`bool`, *optional*):
+            Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
+            more detail.
+        return_dict (`bool`, *optional*):
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
+"""
+
+
+@add_start_docstrings(
+    "The bare LLaMA Model outputting raw hidden-states without any specific head on top.",
+    LLAMA_START_DOCSTRING,
+)
+class OpenMoeModel(OpenMoePreTrainedModel):
+    """
+    Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`LlamaDecoderLayer`]
+
+    Args:
+        config: LlamaConfig
+    """
+
+    def __init__(self, config: LlamaConfig):
+        super().__init__(config)
+        self.padding_idx = config.pad_token_id
+        self.vocab_size = config.vocab_size
+
+        self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx)
+        self.layers = nn.ModuleList(
+            [
+                OpenMoeDecoderLayer(config, moe=True if (i + 1) % config.moe_layer_interval == 0 else False)
+                for i in range(config.num_hidden_layers)
+            ]
+        )
+        self.norm = LlamaRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+
+        self.gradient_checkpointing = False
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.embed_tokens = value
+
+    # Copied from transformers.models.bart.modeling_bart.BartDecoder._prepare_decoder_attention_mask
+    def _prepare_decoder_attention_mask(self, attention_mask, input_shape, inputs_embeds, past_key_values_length):
+        # create causal mask
+        # [bsz, seq_len] -> [bsz, 1, tgt_seq_len, src_seq_len]
+        combined_attention_mask = None
+        if input_shape[-1] > 1:
+            combined_attention_mask = _make_causal_mask(
+                input_shape,
+                inputs_embeds.dtype,
+                device=inputs_embeds.device,
+                past_key_values_length=past_key_values_length,
+            )
+
+        if attention_mask is not None:
+            # [bsz, seq_len] -> [bsz, 1, tgt_seq_len, src_seq_len]
+            expanded_attn_mask = _expand_mask(attention_mask, inputs_embeds.dtype, tgt_len=input_shape[-1]).to(
+                inputs_embeds.device
+            )
+            combined_attention_mask = (
+                expanded_attn_mask if combined_attention_mask is None else expanded_attn_mask + combined_attention_mask
+            )
+
+        return combined_attention_mask
+
+    @add_start_docstrings_to_model_forward(LLAMA_INPUTS_DOCSTRING)
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple, BaseModelOutputWithPast]:
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        # retrieve input_ids and inputs_embeds
+        if input_ids is not None and inputs_embeds is not None:
+            raise ValueError("You cannot specify both decoder_input_ids and decoder_inputs_embeds at the same time")
+        elif input_ids is not None:
+            batch_size, seq_length = input_ids.shape
+        elif inputs_embeds is not None:
+            batch_size, seq_length, _ = inputs_embeds.shape
+        else:
+            raise ValueError("You have to specify either decoder_input_ids or decoder_inputs_embeds")
+
+        seq_length_with_past = seq_length
+        past_key_values_length = 0
+
+        if past_key_values is not None:
+            past_key_values_length = past_key_values[0][0].shape[2]
+            seq_length_with_past = seq_length_with_past + past_key_values_length
+
+        if position_ids is None:
+            device = input_ids.device if input_ids is not None else inputs_embeds.device
+            position_ids = torch.arange(
+                past_key_values_length, seq_length + past_key_values_length, dtype=torch.long, device=device
+            )
+            position_ids = position_ids.unsqueeze(0).view(-1, seq_length)
+        else:
+            position_ids = position_ids.view(-1, seq_length).long()
+
+        if inputs_embeds is None:
+            inputs_embeds = self.embed_tokens(input_ids)
+        # embed positions
+        if attention_mask is None:
+            attention_mask = torch.ones(
+                (batch_size, seq_length_with_past), dtype=torch.bool, device=inputs_embeds.device
+            )
+        attention_mask = self._prepare_decoder_attention_mask(
+            attention_mask, (batch_size, seq_length), inputs_embeds, past_key_values_length
+        )
+
+        hidden_states = inputs_embeds
+
+        if self.gradient_checkpointing and self.training:
+            if use_cache:
+                logger.warning_once(
+                    "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
+                )
+                use_cache = False
+
+        # decoder layers
+        all_hidden_states = () if output_hidden_states else None
+        all_self_attns = () if output_attentions else None
+        next_decoder_cache = () if use_cache else None
+
+        for idx, decoder_layer in enumerate(self.layers):
+            if output_hidden_states:
+                all_hidden_states += (hidden_states,)
+
+            past_key_value = past_key_values[idx] if past_key_values is not None else None
+
+            if self.gradient_checkpointing and self.training:
+
+                def create_custom_forward(module):
+                    def custom_forward(*inputs):
+                        # None for past_key_value
+                        return module(*inputs, output_attentions, None)
+
+                    return custom_forward
+
+                layer_outputs = torch.utils.checkpoint.checkpoint(
+                    create_custom_forward(decoder_layer),
+                    hidden_states,
+                    attention_mask,
+                    position_ids,
+                    None,
+                )
+            else:
+                layer_outputs = decoder_layer(
+                    hidden_states,
+                    attention_mask=attention_mask,
+                    position_ids=position_ids,
+                    past_key_value=past_key_value,
+                    output_attentions=output_attentions,
+                    use_cache=use_cache,
+                )
+
+            hidden_states = layer_outputs[0]
+
+            if use_cache:
+                next_decoder_cache += (layer_outputs[2 if output_attentions else 1],)
+
+            if output_attentions:
+                all_self_attns += (layer_outputs[1],)
+
+        hidden_states = self.norm(hidden_states)
+
+        # add hidden states from the last decoder layer
+        if output_hidden_states:
+            all_hidden_states += (hidden_states,)
+
+        next_cache = next_decoder_cache if use_cache else None
+        if not return_dict:
+            return tuple(v for v in [hidden_states, next_cache, all_hidden_states, all_self_attns] if v is not None)
+        return BaseModelOutputWithPast(
+            last_hidden_state=hidden_states,
+            past_key_values=next_cache,
+            hidden_states=all_hidden_states,
+            attentions=all_self_attns,
+        )
+
+
+class OpenMoeForCausalLM(OpenMoePreTrainedModel):
+    # _tied_weights_keys = ["lm_head.weight"]
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.model = OpenMoeModel(config)
+        self.pretraining_tp = config.pretraining_tp
+        self.vocab_size = config.vocab_size
+        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.model.embed_tokens = value
+
+    def get_output_embeddings(self):
+        return self.lm_head
+
+    def set_output_embeddings(self, new_embeddings):
+        self.lm_head = new_embeddings
+
+    def set_decoder(self, decoder):
+        self.model = decoder
+
+    def get_decoder(self):
+        return self.model
+
+    @add_start_docstrings_to_model_forward(LLAMA_INPUTS_DOCSTRING)
+    @replace_return_docstrings(output_type=CausalLMOutputWithPast, config_class=_CONFIG_FOR_DOC)
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        chunk_head: Optional[bool] = True,
+    ) -> Union[Tuple, CausalLMOutputWithPast]:
+        r"""
+        Args:
+            labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+                Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
+                config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
+                (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
+
+        Returns:
+
+        Example:
+
+        ```python
+        >>> from transformers import AutoTokenizer, LlamaForCausalLM
+
+        >>> model = LlamaForCausalLM.from_pretrained(PATH_TO_CONVERTED_WEIGHTS)
+        >>> tokenizer = AutoTokenizer.from_pretrained(PATH_TO_CONVERTED_TOKENIZER)
+
+        >>> prompt = "Hey, are you conscious? Can you talk to me?"
+        >>> inputs = tokenizer(prompt, return_tensors="pt")
+
+        >>> # Generate
+        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
+        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
+        "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
+        ```"""
+        # reset moe loss
+        MOE_MANAGER.reset_loss()
+
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
+        outputs = self.model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        hidden_states = outputs[0]
+        if self.pretraining_tp > 1:
+            lm_head_slices = self.lm_head.weight.split(self.vocab_size // self.pretraining_tp, dim=0)
+            logits = [F.linear(hidden_states, lm_head_slices[i]) for i in range(self.pretraining_tp)]
+            logits = torch.cat(logits, dim=-1)
+
+        loss = None
+        # if no training, just do forward
+        if labels is None:
+            logits = self.lm_head(hidden_states)
+            logits = logits.float()
+        # the vocab size for openmoe is 30w+
+        # which causes great activation memory in training, up to 20G for one sequence
+        # so we use chunk and checkpoint to reduce memory
+        else:
+            if chunk_head == True:
+
+                def create_custom_forward(module):
+                    def custom_forward(*inputs):
+                        logits = module(inputs[0])
+                        logits = logits.float()
+                        # Shift so that tokens < n predict n
+                        shift_logits = logits[..., :-1, :].contiguous().float()
+                        shift_labels = inputs[1][..., 1:].contiguous()
+                        # Flatten the tokens
+                        loss = self._calculate_loss(shift_logits, shift_labels)
+                        return loss
+
+                    return custom_forward
+
+                aux_loss, z_loss = self._calculate_router_loss()
+                loss = aux_loss + z_loss
+                for batch_idx in range(hidden_states.shape[0]):
+                    loss = loss + torch.utils.checkpoint.checkpoint(
+                        create_custom_forward(self.lm_head),
+                        hidden_states[batch_idx : batch_idx + 1, :],
+                        labels[batch_idx : batch_idx + 1, :],
+                    )
+                logits = None
+            else:
+                logits = self.lm_head(hidden_states)
+                logits = logits.float()
+                # Shift so that tokens < n predict n
+                shift_logits = logits[..., :-1, :].contiguous()
+                shift_labels = labels[..., 1:].contiguous()
+                # Flatten the tokens
+                aux_loss, z_loss = self._calculate_router_loss()
+                loss = aux_loss + z_loss
+                loss = loss + self._calculate_loss(shift_logits, shift_labels)
+
+        if not return_dict:
+            output = (logits,) + outputs[1:]
+            return (loss,) + output if loss is not None else output
+
+        return CausalLMOutputWithPast(
+            loss=loss,
+            logits=logits,
+            past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+    def prepare_inputs_for_generation(
+        self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, **kwargs
+    ):
+        if past_key_values:
+            input_ids = input_ids[:, -1:]
+
+        position_ids = kwargs.get("position_ids", None)
+        if attention_mask is not None and position_ids is None:
+            # create position_ids on the fly for batch generation
+            position_ids = attention_mask.long().cumsum(-1) - 1
+            position_ids.masked_fill_(attention_mask == 0, 1)
+            if past_key_values:
+                position_ids = position_ids[:, -1].unsqueeze(-1)
+
+        # if `inputs_embeds` are passed, we only want to use them in the 1st generation step
+        if inputs_embeds is not None and past_key_values is None:
+            model_inputs = {"inputs_embeds": inputs_embeds}
+        else:
+            model_inputs = {"input_ids": input_ids}
+
+        model_inputs.update(
+            {
+                "position_ids": position_ids,
+                "past_key_values": past_key_values,
+                "use_cache": kwargs.get("use_cache"),
+                "attention_mask": attention_mask,
+            }
+        )
+        return model_inputs
+
+    @staticmethod
+    def _reorder_cache(past_key_values, beam_idx):
+        reordered_past = ()
+        for layer_past in past_key_values:
+            reordered_past += (
+                tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past),
+            )
+        return reordered_past
+
+    def _calculate_router_loss(self, aux_loss: list = None, z_loss: list = None):
+        if aux_loss is None or z_loss is None:
+            aux_loss, z_loss = MOE_MANAGER.get_loss()
+        assert len(aux_loss) == len(z_loss) == self.config.num_hidden_layers // self.config.moe_layer_interval
+        aux_loss = self.config.router_aux_loss_factor * sum(aux_loss) / len(aux_loss)
+        z_loss = self.config.router_z_loss_factor * sum(z_loss) / len(z_loss)
+        return aux_loss, z_loss
+
+    def _calculate_loss(self, logits: torch.Tensor, targets: torch.Tensor) -> torch.Tensor:
+        """Compute cross entropy and entropy for log probs and targets.
+
+        Args:
+            logits: [batch, length, num_classes] float array.
+            targets: categorical targets [batch, length] int array.
+
+        Returns:
+            Tuple of scalar loss.
+        """
+        if len(logits.shape) != len(targets.shape) + 1:
+            raise ValueError(
+                "Incorrect shapes. Got shape %s logits and %s targets" % (str(logits.shape), str(targets.shape))
+            )
+        vocab_size = logits.shape[-1]
+        confidence = 1.0 - self.config.label_smoothing
+        low_confidence = (1.0 - confidence) / (vocab_size - 1)
+        normalizing_constant = -(
+            confidence * math.log(confidence) + (vocab_size - 1) * low_confidence * math.log(low_confidence + 1e-20)
+        )
+
+        # one hot
+        soft_targets = targets[..., None] == torch.arange(vocab_size, device=targets.device).reshape(
+            (1,) * len(targets.shape) + (-1,)
+        )
+        soft_targets = torch.where(
+            soft_targets, torch.full_like(soft_targets, confidence), torch.full_like(soft_targets, low_confidence)
+        )
+        soft_targets = soft_targets.to(torch.float32)
+
+        # cross entropy
+        total_loss = ZLossCrossEntropy.apply(logits, soft_targets, self.config.z_loss_factor)
+        total_loss = total_loss - normalizing_constant
+        total_loss = torch.mean(torch.sum(total_loss, dim=-1), dim=0)
+        return total_loss
+
+
+class ZLossCrossEntropy(torch.autograd.Function):
+    """Computes cross entropy loss with stable custom gradient.
+
+    Computes a stabilized-gradient version of:
+        -jnp.sum(targets * nn.log_softmax(logits), axis=-1)
+
+    If z_loss > 0, then an auxiliary loss equal to z_loss*log(z)^2
+    will be added to the cross entropy loss (z = softmax normalization constant).
+    The two uses of z_loss are:
+    1. To keep the logits from drifting too far from zero, which can cause
+        unacceptable roundoff errors in bfloat16.
+    2. To encourage the logits to be normalized log-probabilities.
+
+    Args:
+        logits: [batch, length, num_classes] float array.
+        targets: categorical one-hot targets [batch, length, num_classes] float
+        array.
+        z_loss: coefficient for auxilliary z-loss loss term.
+
+    Returns:
+        tuple with the total loss and the z_loss, both
+        float arrays with shape [batch, length].
+    """
+
+    @staticmethod
+    def forward(ctx, logits, targets, z_loss):
+        max_logit = torch.max(logits, dim=-1, keepdim=True)[0]
+        shifted = logits - max_logit
+        exp_shifted = torch.exp(shifted)
+        sum_exp = torch.sum(exp_shifted, axis=-1, keepdims=True)
+        sum_exp_log = torch.log(sum_exp)
+        log_softmax = shifted - sum_exp_log
+        loss = -torch.sum(targets * log_softmax, axis=-1)
+        # Add auxilliary z-loss term.
+        log_z = torch.squeeze(sum_exp_log + max_logit, axis=-1)
+        total_z_loss = z_loss * torch.square(log_z)
+        loss += total_z_loss
+        ctx.z_loss = z_loss
+        ctx.save_for_backward(logits, targets, exp_shifted, sum_exp, log_softmax, log_z)
+        return loss
+
+    @staticmethod
+    def backward(ctx, *grad_outputs):
+        assert len(grad_outputs) == 1
+        g = grad_outputs[0]
+        z_loss = ctx.z_loss
+        logits, targets, exp_shifted, sum_exp, log_softmax, log_z = ctx.saved_tensors
+        # z-loss term adds the (2 * z_loss * log_z) factor.
+        deriv = (1 + 2 * z_loss * log_z).unsqueeze(-1) * exp_shifted / sum_exp - targets
+        g_logits = g.unsqueeze(-1) * deriv
+        g_targets = -g.unsqueeze(-1) * log_softmax
+
+        return (
+            g_logits.to(logits.dtype),
+            g_targets.to(targets.dtype),
+            None,
+        )

special_tokens_map.json

@@ -0,0 +1,308 @@
+{
+  "additional_special_tokens": [
+    "<extra_id_0>",
+    "<extra_id_1>",
+    "<extra_id_2>",
+    "<extra_id_3>",
+    "<extra_id_4>",
+    "<extra_id_5>",
+    "<extra_id_6>",
+    "<extra_id_7>",
+    "<extra_id_8>",
+    "<extra_id_9>",
+    "<extra_id_10>",
+    "<extra_id_11>",
+    "<extra_id_12>",
+    "<extra_id_13>",
+    "<extra_id_14>",
+    "<extra_id_15>",
+    "<extra_id_16>",
+    "<extra_id_17>",
+    "<extra_id_18>",
+    "<extra_id_19>",
+    "<extra_id_20>",
+    "<extra_id_21>",
+    "<extra_id_22>",
+    "<extra_id_23>",
+    "<extra_id_24>",
+    "<extra_id_25>",
+    "<extra_id_26>",
+    "<extra_id_27>",
+    "<extra_id_28>",
+    "<extra_id_29>",
+    "<extra_id_30>",
+    "<extra_id_31>",
+    "<extra_id_32>",
+    "<extra_id_33>",
+    "<extra_id_34>",
+    "<extra_id_35>",
+    "<extra_id_36>",
+    "<extra_id_37>",
+    "<extra_id_38>",
+    "<extra_id_39>",
+    "<extra_id_40>",
+    "<extra_id_41>",
+    "<extra_id_42>",
+    "<extra_id_43>",
+    "<extra_id_44>",
+    "<extra_id_45>",
+    "<extra_id_46>",
+    "<extra_id_47>",
+    "<extra_id_48>",
+    "<extra_id_49>",
+    "<extra_id_50>",
+    "<extra_id_51>",
+    "<extra_id_52>",
+    "<extra_id_53>",
+    "<extra_id_54>",
+    "<extra_id_55>",
+    "<extra_id_56>",
+    "<extra_id_57>",
+    "<extra_id_58>",
+    "<extra_id_59>",
+    "<extra_id_60>",
+    "<extra_id_61>",
+    "<extra_id_62>",
+    "<extra_id_63>",
+    "<extra_id_64>",
+    "<extra_id_65>",
+    "<extra_id_66>",
+    "<extra_id_67>",
+    "<extra_id_68>",
+    "<extra_id_69>",
+    "<extra_id_70>",
+    "<extra_id_71>",
+    "<extra_id_72>",
+    "<extra_id_73>",
+    "<extra_id_74>",
+    "<extra_id_75>",
+    "<extra_id_76>",
+    "<extra_id_77>",
+    "<extra_id_78>",
+    "<extra_id_79>",
+    "<extra_id_80>",
+    "<extra_id_81>",
+    "<extra_id_82>",
+    "<extra_id_83>",
+    "<extra_id_84>",
+    "<extra_id_85>",
+    "<extra_id_86>",
+    "<extra_id_87>",
+    "<extra_id_88>",
+    "<extra_id_89>",
+    "<extra_id_90>",
+    "<extra_id_91>",
+    "<extra_id_92>",
+    "<extra_id_93>",
+    "<extra_id_94>",
+    "<extra_id_95>",
+    "<extra_id_96>",
+    "<extra_id_97>",
+    "<extra_id_98>",
+    "<extra_id_99>",
+    "<extra_id_100>",
+    "<extra_id_101>",
+    "<extra_id_102>",
+    "<extra_id_103>",
+    "<extra_id_104>",
+    "<extra_id_105>",
+    "<extra_id_106>",
+    "<extra_id_107>",
+    "<extra_id_108>",
+    "<extra_id_109>",
+    "<extra_id_110>",
+    "<extra_id_111>",
+    "<extra_id_112>",
+    "<extra_id_113>",
+    "<extra_id_114>",
+    "<extra_id_115>",
+    "<extra_id_116>",
+    "<extra_id_117>",
+    "<extra_id_118>",
+    "<extra_id_119>",
+    "<extra_id_120>",
+    "<extra_id_121>",
+    "<extra_id_122>",
+    "<extra_id_123>",
+    "<extra_id_124>",
+    "<extra_id_125>",
+    "<extra_id_126>",
+    "<extra_id_127>",
+    "<extra_id_128>",
+    "<extra_id_129>",
+    "<extra_id_130>",
+    "<extra_id_131>",
+    "<extra_id_132>",
+    "<extra_id_133>",
+    "<extra_id_134>",
+    "<extra_id_135>",
+    "<extra_id_136>",
+    "<extra_id_137>",
+    "<extra_id_138>",
+    "<extra_id_139>",
+    "<extra_id_140>",
+    "<extra_id_141>",
+    "<extra_id_142>",
+    "<extra_id_143>",
+    "<extra_id_144>",
+    "<extra_id_145>",
+    "<extra_id_146>",
+    "<extra_id_147>",
+    "<extra_id_148>",
+    "<extra_id_149>",
+    "<extra_id_150>",
+    "<extra_id_151>",
+    "<extra_id_152>",
+    "<extra_id_153>",
+    "<extra_id_154>",
+    "<extra_id_155>",
+    "<extra_id_156>",
+    "<extra_id_157>",
+    "<extra_id_158>",
+    "<extra_id_159>",
+    "<extra_id_160>",
+    "<extra_id_161>",
+    "<extra_id_162>",
+    "<extra_id_163>",
+    "<extra_id_164>",
+    "<extra_id_165>",
+    "<extra_id_166>",
+    "<extra_id_167>",
+    "<extra_id_168>",
+    "<extra_id_169>",
+    "<extra_id_170>",
+    "<extra_id_171>",
+    "<extra_id_172>",
+    "<extra_id_173>",
+    "<extra_id_174>",
+    "<extra_id_175>",
+    "<extra_id_176>",
+    "<extra_id_177>",
+    "<extra_id_178>",
+    "<extra_id_179>",
+    "<extra_id_180>",
+    "<extra_id_181>",
+    "<extra_id_182>",
+    "<extra_id_183>",
+    "<extra_id_184>",
+    "<extra_id_185>",
+    "<extra_id_186>",
+    "<extra_id_187>",
+    "<extra_id_188>",
+    "<extra_id_189>",
+    "<extra_id_190>",
+    "<extra_id_191>",
+    "<extra_id_192>",
+    "<extra_id_193>",
+    "<extra_id_194>",
+    "<extra_id_195>",
+    "<extra_id_196>",
+    "<extra_id_197>",
+    "<extra_id_198>",
+    "<extra_id_199>",
+    "<extra_id_200>",
+    "<extra_id_201>",
+    "<extra_id_202>",
+    "<extra_id_203>",
+    "<extra_id_204>",
+    "<extra_id_205>",
+    "<extra_id_206>",
+    "<extra_id_207>",
+    "<extra_id_208>",
+    "<extra_id_209>",
+    "<extra_id_210>",
+    "<extra_id_211>",
+    "<extra_id_212>",
+    "<extra_id_213>",
+    "<extra_id_214>",
+    "<extra_id_215>",
+    "<extra_id_216>",
+    "<extra_id_217>",
+    "<extra_id_218>",
+    "<extra_id_219>",
+    "<extra_id_220>",
+    "<extra_id_221>",
+    "<extra_id_222>",
+    "<extra_id_223>",
+    "<extra_id_224>",
+    "<extra_id_225>",
+    "<extra_id_226>",
+    "<extra_id_227>",
+    "<extra_id_228>",
+    "<extra_id_229>",
+    "<extra_id_230>",
+    "<extra_id_231>",
+    "<extra_id_232>",
+    "<extra_id_233>",
+    "<extra_id_234>",
+    "<extra_id_235>",
+    "<extra_id_236>",
+    "<extra_id_237>",
+    "<extra_id_238>",
+    "<extra_id_239>",
+    "<extra_id_240>",
+    "<extra_id_241>",
+    "<extra_id_242>",
+    "<extra_id_243>",
+    "<extra_id_244>",
+    "<extra_id_245>",
+    "<extra_id_246>",
+    "<extra_id_247>",
+    "<extra_id_248>",
+    "<extra_id_249>",
+    "<extra_id_250>",
+    "<extra_id_251>",
+    "<extra_id_252>",
+    "<extra_id_253>",
+    "<extra_id_254>",
+    "<extra_id_255>",
+    "<extra_id_256>",
+    "<extra_id_257>",
+    "<extra_id_258>",
+    "<extra_id_259>",
+    "<extra_id_260>",
+    "<extra_id_261>",
+    "<extra_id_262>",
+    "<extra_id_263>",
+    "<extra_id_264>",
+    "<extra_id_265>",
+    "<extra_id_266>",
+    "<extra_id_267>",
+    "<extra_id_268>",
+    "<extra_id_269>",
+    "<extra_id_270>",
+    "<extra_id_271>",
+    "<extra_id_272>",
+    "<extra_id_273>",
+    "<extra_id_274>",
+    "<extra_id_275>",
+    "<extra_id_276>",
+    "<extra_id_277>",
+    "<extra_id_278>",
+    "<extra_id_279>",
+    "<extra_id_280>",
+    "<extra_id_281>",
+    "<extra_id_282>",
+    "<extra_id_283>",
+    "<extra_id_284>",
+    "<extra_id_285>",
+    "<extra_id_286>",
+    "<extra_id_287>",
+    "<extra_id_288>",
+    "<extra_id_289>",
+    "<extra_id_290>",
+    "<extra_id_291>",
+    "<extra_id_292>",
+    "<extra_id_293>",
+    "<extra_id_294>",
+    "<extra_id_295>",
+    "<extra_id_296>",
+    "<extra_id_297>",
+    "<extra_id_298>",
+    "<extra_id_299>"
+  ],
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

spiece.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3909a67b780650b35cf529ac782ad2b6b26e6d1f849d3fbb6a872905f452458
+size 4548313

tokenization_openmoe.py

@@ -0,0 +1,22 @@
+from transformers import T5Tokenizer
+from typing import List, Optional, Tuple, Union
+
+class OpenMoeTokenizer(T5Tokenizer):
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.padding_side = 'left'
+        self.add_bos_token = True
+        self.add_eos_token = False
+
+    def build_inputs_with_special_tokens(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        if self.add_eos_token:
+            token_ids_0 = self._add_eos_if_not_present(token_ids_0)
+        if self.add_bos_token:
+            token_ids_0 = [self.pad_token_id] + token_ids_0
+        if token_ids_1 is None:
+            return token_ids_0
+        else:
+            token_ids_1 = self._add_eos_if_not_present(token_ids_1)
+            return token_ids_0 + token_ids_1

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af904105ce1071b1202bba0059a841f4a7b85b48b6ec179c4948e3483476e0dd
+size 16853013

tokenizer_config.json

@@ -0,0 +1,2757 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "1": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "2": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256000": {
+      "content": "<extra_id_299>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256001": {
+      "content": "<extra_id_298>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256002": {
+      "content": "<extra_id_297>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256003": {
+      "content": "<extra_id_296>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256004": {
+      "content": "<extra_id_295>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256005": {
+      "content": "<extra_id_294>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256006": {
+      "content": "<extra_id_293>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256007": {
+      "content": "<extra_id_292>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256008": {
+      "content": "<extra_id_291>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256009": {
+      "content": "<extra_id_290>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256010": {
+      "content": "<extra_id_289>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256011": {
+      "content": "<extra_id_288>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256012": {
+      "content": "<extra_id_287>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256013": {
+      "content": "<extra_id_286>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256014": {
+      "content": "<extra_id_285>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256015": {
+      "content": "<extra_id_284>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256016": {
+      "content": "<extra_id_283>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256017": {
+      "content": "<extra_id_282>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256018": {
+      "content": "<extra_id_281>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256019": {
+      "content": "<extra_id_280>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256020": {
+      "content": "<extra_id_279>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256021": {
+      "content": "<extra_id_278>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256022": {
+      "content": "<extra_id_277>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256023": {
+      "content": "<extra_id_276>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256024": {
+      "content": "<extra_id_275>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256025": {
+      "content": "<extra_id_274>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256026": {
+      "content": "<extra_id_273>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256027": {
+      "content": "<extra_id_272>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256028": {
+      "content": "<extra_id_271>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256029": {
+      "content": "<extra_id_270>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256030": {
+      "content": "<extra_id_269>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256031": {
+      "content": "<extra_id_268>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256032": {
+      "content": "<extra_id_267>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256033": {
+      "content": "<extra_id_266>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256034": {
+      "content": "<extra_id_265>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256035": {
+      "content": "<extra_id_264>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256036": {
+      "content": "<extra_id_263>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256037": {
+      "content": "<extra_id_262>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256038": {
+      "content": "<extra_id_261>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256039": {
+      "content": "<extra_id_260>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256040": {
+      "content": "<extra_id_259>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256041": {
+      "content": "<extra_id_258>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256042": {
+      "content": "<extra_id_257>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256043": {
+      "content": "<extra_id_256>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256044": {
+      "content": "<extra_id_255>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256045": {
+      "content": "<extra_id_254>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256046": {
+      "content": "<extra_id_253>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256047": {
+      "content": "<extra_id_252>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256048": {
+      "content": "<extra_id_251>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256049": {
+      "content": "<extra_id_250>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256050": {
+      "content": "<extra_id_249>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256051": {
+      "content": "<extra_id_248>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256052": {
+      "content": "<extra_id_247>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256053": {
+      "content": "<extra_id_246>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256054": {
+      "content": "<extra_id_245>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256055": {
+      "content": "<extra_id_244>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256056": {
+      "content": "<extra_id_243>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256057": {
+      "content": "<extra_id_242>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256058": {
+      "content": "<extra_id_241>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256059": {
+      "content": "<extra_id_240>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256060": {
+      "content": "<extra_id_239>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256061": {
+      "content": "<extra_id_238>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256062": {
+      "content": "<extra_id_237>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256063": {
+      "content": "<extra_id_236>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256064": {
+      "content": "<extra_id_235>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256065": {
+      "content": "<extra_id_234>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256066": {
+      "content": "<extra_id_233>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256067": {
+      "content": "<extra_id_232>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256068": {
+      "content": "<extra_id_231>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256069": {
+      "content": "<extra_id_230>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256070": {
+      "content": "<extra_id_229>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256071": {
+      "content": "<extra_id_228>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256072": {
+      "content": "<extra_id_227>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256073": {
+      "content": "<extra_id_226>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256074": {
+      "content": "<extra_id_225>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256075": {
+      "content": "<extra_id_224>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256076": {
+      "content": "<extra_id_223>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256077": {
+      "content": "<extra_id_222>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256078": {
+      "content": "<extra_id_221>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256079": {
+      "content": "<extra_id_220>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256080": {
+      "content": "<extra_id_219>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256081": {
+      "content": "<extra_id_218>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256082": {
+      "content": "<extra_id_217>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256083": {
+      "content": "<extra_id_216>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256084": {
+      "content": "<extra_id_215>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256085": {
+      "content": "<extra_id_214>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256086": {
+      "content": "<extra_id_213>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256087": {
+      "content": "<extra_id_212>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256088": {
+      "content": "<extra_id_211>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256089": {
+      "content": "<extra_id_210>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256090": {
+      "content": "<extra_id_209>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256091": {
+      "content": "<extra_id_208>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256092": {
+      "content": "<extra_id_207>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256093": {
+      "content": "<extra_id_206>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256094": {
+      "content": "<extra_id_205>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256095": {
+      "content": "<extra_id_204>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256096": {
+      "content": "<extra_id_203>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256097": {
+      "content": "<extra_id_202>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256098": {
+      "content": "<extra_id_201>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256099": {
+      "content": "<extra_id_200>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256100": {
+      "content": "<extra_id_199>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256101": {
+      "content": "<extra_id_198>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256102": {
+      "content": "<extra_id_197>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256103": {
+      "content": "<extra_id_196>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256104": {
+      "content": "<extra_id_195>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256105": {
+      "content": "<extra_id_194>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256106": {
+      "content": "<extra_id_193>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256107": {
+      "content": "<extra_id_192>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256108": {
+      "content": "<extra_id_191>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256109": {
+      "content": "<extra_id_190>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256110": {
+      "content": "<extra_id_189>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256111": {
+      "content": "<extra_id_188>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256112": {
+      "content": "<extra_id_187>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256113": {
+      "content": "<extra_id_186>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256114": {
+      "content": "<extra_id_185>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256115": {
+      "content": "<extra_id_184>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256116": {
+      "content": "<extra_id_183>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256117": {
+      "content": "<extra_id_182>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256118": {
+      "content": "<extra_id_181>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256119": {
+      "content": "<extra_id_180>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256120": {
+      "content": "<extra_id_179>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256121": {
+      "content": "<extra_id_178>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256122": {
+      "content": "<extra_id_177>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256123": {
+      "content": "<extra_id_176>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256124": {
+      "content": "<extra_id_175>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256125": {
+      "content": "<extra_id_174>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256126": {
+      "content": "<extra_id_173>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256127": {
+      "content": "<extra_id_172>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256128": {
+      "content": "<extra_id_171>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256129": {
+      "content": "<extra_id_170>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256130": {
+      "content": "<extra_id_169>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256131": {
+      "content": "<extra_id_168>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256132": {
+      "content": "<extra_id_167>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256133": {
+      "content": "<extra_id_166>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256134": {
+      "content": "<extra_id_165>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256135": {
+      "content": "<extra_id_164>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256136": {
+      "content": "<extra_id_163>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256137": {
+      "content": "<extra_id_162>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256138": {
+      "content": "<extra_id_161>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256139": {
+      "content": "<extra_id_160>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256140": {
+      "content": "<extra_id_159>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256141": {
+      "content": "<extra_id_158>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256142": {
+      "content": "<extra_id_157>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256143": {
+      "content": "<extra_id_156>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256144": {
+      "content": "<extra_id_155>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256145": {
+      "content": "<extra_id_154>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256146": {
+      "content": "<extra_id_153>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256147": {
+      "content": "<extra_id_152>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256148": {
+      "content": "<extra_id_151>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256149": {
+      "content": "<extra_id_150>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256150": {
+      "content": "<extra_id_149>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256151": {
+      "content": "<extra_id_148>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256152": {
+      "content": "<extra_id_147>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256153": {
+      "content": "<extra_id_146>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256154": {
+      "content": "<extra_id_145>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256155": {
+      "content": "<extra_id_144>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256156": {
+      "content": "<extra_id_143>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256157": {
+      "content": "<extra_id_142>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256158": {
+      "content": "<extra_id_141>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256159": {
+      "content": "<extra_id_140>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256160": {
+      "content": "<extra_id_139>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256161": {
+      "content": "<extra_id_138>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256162": {
+      "content": "<extra_id_137>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256163": {
+      "content": "<extra_id_136>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256164": {
+      "content": "<extra_id_135>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256165": {
+      "content": "<extra_id_134>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256166": {
+      "content": "<extra_id_133>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256167": {
+      "content": "<extra_id_132>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256168": {
+      "content": "<extra_id_131>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256169": {
+      "content": "<extra_id_130>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256170": {
+      "content": "<extra_id_129>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256171": {
+      "content": "<extra_id_128>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256172": {
+      "content": "<extra_id_127>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256173": {
+      "content": "<extra_id_126>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256174": {
+      "content": "<extra_id_125>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256175": {
+      "content": "<extra_id_124>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256176": {
+      "content": "<extra_id_123>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256177": {
+      "content": "<extra_id_122>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256178": {
+      "content": "<extra_id_121>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256179": {
+      "content": "<extra_id_120>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256180": {
+      "content": "<extra_id_119>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256181": {
+      "content": "<extra_id_118>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256182": {
+      "content": "<extra_id_117>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256183": {
+      "content": "<extra_id_116>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256184": {
+      "content": "<extra_id_115>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256185": {
+      "content": "<extra_id_114>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256186": {
+      "content": "<extra_id_113>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256187": {
+      "content": "<extra_id_112>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256188": {
+      "content": "<extra_id_111>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256189": {
+      "content": "<extra_id_110>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256190": {
+      "content": "<extra_id_109>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256191": {
+      "content": "<extra_id_108>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256192": {
+      "content": "<extra_id_107>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256193": {
+      "content": "<extra_id_106>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256194": {
+      "content": "<extra_id_105>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256195": {
+      "content": "<extra_id_104>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256196": {
+      "content": "<extra_id_103>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256197": {
+      "content": "<extra_id_102>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256198": {
+      "content": "<extra_id_101>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256199": {
+      "content": "<extra_id_100>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256200": {
+      "content": "<extra_id_99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256201": {
+      "content": "<extra_id_98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256202": {
+      "content": "<extra_id_97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256203": {
+      "content": "<extra_id_96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256204": {
+      "content": "<extra_id_95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256205": {
+      "content": "<extra_id_94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256206": {
+      "content": "<extra_id_93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256207": {
+      "content": "<extra_id_92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256208": {
+      "content": "<extra_id_91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256209": {
+      "content": "<extra_id_90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256210": {
+      "content": "<extra_id_89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256211": {
+      "content": "<extra_id_88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256212": {
+      "content": "<extra_id_87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256213": {
+      "content": "<extra_id_86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256214": {
+      "content": "<extra_id_85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256215": {
+      "content": "<extra_id_84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256216": {
+      "content": "<extra_id_83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256217": {
+      "content": "<extra_id_82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256218": {
+      "content": "<extra_id_81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256219": {
+      "content": "<extra_id_80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256220": {
+      "content": "<extra_id_79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256221": {
+      "content": "<extra_id_78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256222": {
+      "content": "<extra_id_77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256223": {
+      "content": "<extra_id_76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256224": {
+      "content": "<extra_id_75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256225": {
+      "content": "<extra_id_74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256226": {
+      "content": "<extra_id_73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256227": {
+      "content": "<extra_id_72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256228": {
+      "content": "<extra_id_71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256229": {
+      "content": "<extra_id_70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256230": {
+      "content": "<extra_id_69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256231": {
+      "content": "<extra_id_68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256232": {
+      "content": "<extra_id_67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256233": {
+      "content": "<extra_id_66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256234": {
+      "content": "<extra_id_65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256235": {
+      "content": "<extra_id_64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256236": {
+      "content": "<extra_id_63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256237": {
+      "content": "<extra_id_62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256238": {
+      "content": "<extra_id_61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256239": {
+      "content": "<extra_id_60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256240": {
+      "content": "<extra_id_59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256241": {
+      "content": "<extra_id_58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256242": {
+      "content": "<extra_id_57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256243": {
+      "content": "<extra_id_56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256244": {
+      "content": "<extra_id_55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256245": {
+      "content": "<extra_id_54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256246": {
+      "content": "<extra_id_53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256247": {
+      "content": "<extra_id_52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256248": {
+      "content": "<extra_id_51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256249": {
+      "content": "<extra_id_50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256250": {
+      "content": "<extra_id_49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256251": {
+      "content": "<extra_id_48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256252": {
+      "content": "<extra_id_47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256253": {
+      "content": "<extra_id_46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256254": {
+      "content": "<extra_id_45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256255": {
+      "content": "<extra_id_44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256256": {
+      "content": "<extra_id_43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256257": {
+      "content": "<extra_id_42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256258": {
+      "content": "<extra_id_41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256259": {
+      "content": "<extra_id_40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256260": {
+      "content": "<extra_id_39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256261": {
+      "content": "<extra_id_38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256262": {
+      "content": "<extra_id_37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256263": {
+      "content": "<extra_id_36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256264": {
+      "content": "<extra_id_35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256265": {
+      "content": "<extra_id_34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256266": {
+      "content": "<extra_id_33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256267": {
+      "content": "<extra_id_32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256268": {
+      "content": "<extra_id_31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256269": {
+      "content": "<extra_id_30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256270": {
+      "content": "<extra_id_29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256271": {
+      "content": "<extra_id_28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256272": {
+      "content": "<extra_id_27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256273": {
+      "content": "<extra_id_26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256274": {
+      "content": "<extra_id_25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256275": {
+      "content": "<extra_id_24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256276": {
+      "content": "<extra_id_23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256277": {
+      "content": "<extra_id_22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256278": {
+      "content": "<extra_id_21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256279": {
+      "content": "<extra_id_20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256280": {
+      "content": "<extra_id_19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256281": {
+      "content": "<extra_id_18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256282": {
+      "content": "<extra_id_17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256283": {
+      "content": "<extra_id_16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256284": {
+      "content": "<extra_id_15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256285": {
+      "content": "<extra_id_14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256286": {
+      "content": "<extra_id_13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256287": {
+      "content": "<extra_id_12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256288": {
+      "content": "<extra_id_11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256289": {
+      "content": "<extra_id_10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256290": {
+      "content": "<extra_id_9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256291": {
+      "content": "<extra_id_8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256292": {
+      "content": "<extra_id_7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256293": {
+      "content": "<extra_id_6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256294": {
+      "content": "<extra_id_5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256295": {
+      "content": "<extra_id_4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256296": {
+      "content": "<extra_id_3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256297": {
+      "content": "<extra_id_2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256298": {
+      "content": "<extra_id_1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "256299": {
+      "content": "<extra_id_0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<extra_id_0>",
+    "<extra_id_1>",
+    "<extra_id_2>",
+    "<extra_id_3>",
+    "<extra_id_4>",
+    "<extra_id_5>",
+    "<extra_id_6>",
+    "<extra_id_7>",
+    "<extra_id_8>",
+    "<extra_id_9>",
+    "<extra_id_10>",
+    "<extra_id_11>",
+    "<extra_id_12>",
+    "<extra_id_13>",
+    "<extra_id_14>",
+    "<extra_id_15>",
+    "<extra_id_16>",
+    "<extra_id_17>",
+    "<extra_id_18>",
+    "<extra_id_19>",
+    "<extra_id_20>",
+    "<extra_id_21>",
+    "<extra_id_22>",
+    "<extra_id_23>",
+    "<extra_id_24>",
+    "<extra_id_25>",
+    "<extra_id_26>",
+    "<extra_id_27>",
+    "<extra_id_28>",
+    "<extra_id_29>",
+    "<extra_id_30>",
+    "<extra_id_31>",
+    "<extra_id_32>",
+    "<extra_id_33>",
+    "<extra_id_34>",
+    "<extra_id_35>",
+    "<extra_id_36>",
+    "<extra_id_37>",
+    "<extra_id_38>",
+    "<extra_id_39>",
+    "<extra_id_40>",
+    "<extra_id_41>",
+    "<extra_id_42>",
+    "<extra_id_43>",
+    "<extra_id_44>",
+    "<extra_id_45>",
+    "<extra_id_46>",
+    "<extra_id_47>",
+    "<extra_id_48>",
+    "<extra_id_49>",
+    "<extra_id_50>",
+    "<extra_id_51>",
+    "<extra_id_52>",
+    "<extra_id_53>",
+    "<extra_id_54>",
+    "<extra_id_55>",
+    "<extra_id_56>",
+    "<extra_id_57>",
+    "<extra_id_58>",
+    "<extra_id_59>",
+    "<extra_id_60>",
+    "<extra_id_61>",
+    "<extra_id_62>",
+    "<extra_id_63>",
+    "<extra_id_64>",
+    "<extra_id_65>",
+    "<extra_id_66>",
+    "<extra_id_67>",
+    "<extra_id_68>",
+    "<extra_id_69>",
+    "<extra_id_70>",
+    "<extra_id_71>",
+    "<extra_id_72>",
+    "<extra_id_73>",
+    "<extra_id_74>",
+    "<extra_id_75>",
+    "<extra_id_76>",
+    "<extra_id_77>",
+    "<extra_id_78>",
+    "<extra_id_79>",
+    "<extra_id_80>",
+    "<extra_id_81>",
+    "<extra_id_82>",
+    "<extra_id_83>",
+    "<extra_id_84>",
+    "<extra_id_85>",
+    "<extra_id_86>",
+    "<extra_id_87>",
+    "<extra_id_88>",
+    "<extra_id_89>",
+    "<extra_id_90>",
+    "<extra_id_91>",
+    "<extra_id_92>",
+    "<extra_id_93>",
+    "<extra_id_94>",
+    "<extra_id_95>",
+    "<extra_id_96>",
+    "<extra_id_97>",
+    "<extra_id_98>",
+    "<extra_id_99>",
+    "<extra_id_100>",
+    "<extra_id_101>",
+    "<extra_id_102>",
+    "<extra_id_103>",
+    "<extra_id_104>",
+    "<extra_id_105>",
+    "<extra_id_106>",
+    "<extra_id_107>",
+    "<extra_id_108>",
+    "<extra_id_109>",
+    "<extra_id_110>",
+    "<extra_id_111>",
+    "<extra_id_112>",
+    "<extra_id_113>",
+    "<extra_id_114>",
+    "<extra_id_115>",
+    "<extra_id_116>",
+    "<extra_id_117>",
+    "<extra_id_118>",
+    "<extra_id_119>",
+    "<extra_id_120>",
+    "<extra_id_121>",
+    "<extra_id_122>",
+    "<extra_id_123>",
+    "<extra_id_124>",
+    "<extra_id_125>",
+    "<extra_id_126>",
+    "<extra_id_127>",
+    "<extra_id_128>",
+    "<extra_id_129>",
+    "<extra_id_130>",
+    "<extra_id_131>",
+    "<extra_id_132>",
+    "<extra_id_133>",
+    "<extra_id_134>",
+    "<extra_id_135>",
+    "<extra_id_136>",
+    "<extra_id_137>",
+    "<extra_id_138>",
+    "<extra_id_139>",
+    "<extra_id_140>",
+    "<extra_id_141>",
+    "<extra_id_142>",
+    "<extra_id_143>",
+    "<extra_id_144>",
+    "<extra_id_145>",
+    "<extra_id_146>",
+    "<extra_id_147>",
+    "<extra_id_148>",
+    "<extra_id_149>",
+    "<extra_id_150>",
+    "<extra_id_151>",
+    "<extra_id_152>",
+    "<extra_id_153>",
+    "<extra_id_154>",
+    "<extra_id_155>",
+    "<extra_id_156>",
+    "<extra_id_157>",
+    "<extra_id_158>",
+    "<extra_id_159>",
+    "<extra_id_160>",
+    "<extra_id_161>",
+    "<extra_id_162>",
+    "<extra_id_163>",
+    "<extra_id_164>",
+    "<extra_id_165>",
+    "<extra_id_166>",
+    "<extra_id_167>",
+    "<extra_id_168>",
+    "<extra_id_169>",
+    "<extra_id_170>",
+    "<extra_id_171>",
+    "<extra_id_172>",
+    "<extra_id_173>",
+    "<extra_id_174>",
+    "<extra_id_175>",
+    "<extra_id_176>",
+    "<extra_id_177>",
+    "<extra_id_178>",
+    "<extra_id_179>",
+    "<extra_id_180>",
+    "<extra_id_181>",
+    "<extra_id_182>",
+    "<extra_id_183>",
+    "<extra_id_184>",
+    "<extra_id_185>",
+    "<extra_id_186>",
+    "<extra_id_187>",
+    "<extra_id_188>",
+    "<extra_id_189>",
+    "<extra_id_190>",
+    "<extra_id_191>",
+    "<extra_id_192>",
+    "<extra_id_193>",
+    "<extra_id_194>",
+    "<extra_id_195>",
+    "<extra_id_196>",
+    "<extra_id_197>",
+    "<extra_id_198>",
+    "<extra_id_199>",
+    "<extra_id_200>",
+    "<extra_id_201>",
+    "<extra_id_202>",
+    "<extra_id_203>",
+    "<extra_id_204>",
+    "<extra_id_205>",
+    "<extra_id_206>",
+    "<extra_id_207>",
+    "<extra_id_208>",
+    "<extra_id_209>",
+    "<extra_id_210>",
+    "<extra_id_211>",
+    "<extra_id_212>",
+    "<extra_id_213>",
+    "<extra_id_214>",
+    "<extra_id_215>",
+    "<extra_id_216>",
+    "<extra_id_217>",
+    "<extra_id_218>",
+    "<extra_id_219>",
+    "<extra_id_220>",
+    "<extra_id_221>",
+    "<extra_id_222>",
+    "<extra_id_223>",
+    "<extra_id_224>",
+    "<extra_id_225>",
+    "<extra_id_226>",
+    "<extra_id_227>",
+    "<extra_id_228>",
+    "<extra_id_229>",
+    "<extra_id_230>",
+    "<extra_id_231>",
+    "<extra_id_232>",
+    "<extra_id_233>",
+    "<extra_id_234>",
+    "<extra_id_235>",
+    "<extra_id_236>",
+    "<extra_id_237>",
+    "<extra_id_238>",
+    "<extra_id_239>",
+    "<extra_id_240>",
+    "<extra_id_241>",
+    "<extra_id_242>",
+    "<extra_id_243>",
+    "<extra_id_244>",
+    "<extra_id_245>",
+    "<extra_id_246>",
+    "<extra_id_247>",
+    "<extra_id_248>",
+    "<extra_id_249>",
+    "<extra_id_250>",
+    "<extra_id_251>",
+    "<extra_id_252>",
+    "<extra_id_253>",
+    "<extra_id_254>",
+    "<extra_id_255>",
+    "<extra_id_256>",
+    "<extra_id_257>",
+    "<extra_id_258>",
+    "<extra_id_259>",
+    "<extra_id_260>",
+    "<extra_id_261>",
+    "<extra_id_262>",
+    "<extra_id_263>",
+    "<extra_id_264>",
+    "<extra_id_265>",
+    "<extra_id_266>",
+    "<extra_id_267>",
+    "<extra_id_268>",
+    "<extra_id_269>",
+    "<extra_id_270>",
+    "<extra_id_271>",
+    "<extra_id_272>",
+    "<extra_id_273>",
+    "<extra_id_274>",
+    "<extra_id_275>",
+    "<extra_id_276>",
+    "<extra_id_277>",
+    "<extra_id_278>",
+    "<extra_id_279>",
+    "<extra_id_280>",
+    "<extra_id_281>",
+    "<extra_id_282>",
+    "<extra_id_283>",
+    "<extra_id_284>",
+    "<extra_id_285>",
+    "<extra_id_286>",
+    "<extra_id_287>",
+    "<extra_id_288>",
+    "<extra_id_289>",
+    "<extra_id_290>",
+    "<extra_id_291>",
+    "<extra_id_292>",
+    "<extra_id_293>",
+    "<extra_id_294>",
+    "<extra_id_295>",
+    "<extra_id_296>",
+    "<extra_id_297>",
+    "<extra_id_298>",
+    "<extra_id_299>"
+  ],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "</s>",
+  "extra_ids": 300,
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "OpenMoeTokenizer",
+  "trust_remote_code": true,
+  "unk_token": "<unk>",
+  "verbose": false,
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_openmoe.OpenMoeTokenizer",
+      null
+      ]
+  }
+}