ptrdvn/Karasu-Mixtral-8x22B-v0.1-lora-adapters

These are the QLoRA adapters for training lightblue/Karasu-Mixtral-8x22B-v0.1. There are also 4 checkpoints from training.

See axolotl config

axolotl version: 0.4.0

base_model: mistral-community/Mixtral-8x22B-v0.1
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
trust_remote_code: true

load_in_8bit: false
load_in_4bit: true
strict: false

datasets:
  - path: lightblue/gpt4_conversations_multilingual
    type: sharegpt
    conversation: mistral
dataset_prepared_path: ./prepared_dataset_2048-multiling
val_set_size: 0
output_dir: ./qlora-out-2048-multiling

## You can optionally freeze the entire model and unfreeze a subset of parameters
unfrozen_parameters:
#  - ^lm_head.weight$
#  - ^model.embed_tokens.weight$[:32000]
#  - model.layers.2[0-9]+.block_sparse_moe.gate
#  - model.layers.2[0-9]+.block_sparse_moe.experts
#  - model.layers.3[0-9]+.block_sparse_moe.gate
#  - model.layers.3[0-9]+.block_sparse_moe.experts

model_config:
  output_router_logits: true

adapter: qlora
lora_model_dir:

sequence_len: 2048
sample_packing: true
pad_to_sequence_len: true

lora_r: 16
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:
#lora_target_modules:
#  - gate
#  - q_proj
#  - k_proj
#  - v_proj
#  - o_proj
#  - w1
#  - w2
#  - w3

gradient_accumulation_steps: 2
micro_batch_size: 1
num_epochs: 1
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002

use_wandb: true
wandb_project: axolotl
wandb_entity: peterd
wandb_name: mixtral_8x22b_test

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_steps: 10
evals_per_epoch: 0
eval_table_size:
eval_max_new_tokens: 128
saves_per_epoch: 5
debug:
deepspeed: /workspace/axolotl/deepspeed_configs/zero2.json
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:

qlora-out-2048-multiling

This model is a fine-tuned version of mistral-community/Mixtral-8x22B-v0.1 on the None dataset.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• total_eval_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 1

Training results

Framework versions

• PEFT 0.10.0
• Transformers 4.40.0.dev0
• Pytorch 2.1.2+cu121
• Datasets 2.18.0
• Tokenizers 0.15.0