Upload folder using huggingface_hub (#1)

- 77d1179c6b657880313d045d137679962b388983d43bbc541a31928ccbe7f805 (d4fe00dee14f552b4dbe4a0c8edf96080ded689e)
- e89ccb677fad7cdb7b90a0491afedbaa2759ef79e24b03f1facddc6498c2a323 (bf584d51f0573907f968ecd9300d5534b86723e6)
- 6ad4ca878787f2072213f7cca29bb9fb806c75261f22e005c029de9cd1a05292 (a9cdf2026bc632d358eb3481f0db928cb0bf5e7e)

README.md

@@ -0,0 +1,34 @@
+---
+base_model: mistralai/Mamba-Codestral-7B-v0.1
+license: apache-2.0
+tags:
+- mlx
+extra_gated_description: If you want to learn more about how we process your personal
+  data, please read our <a href="https://mistral.ai/terms/">Privacy Policy</a>.
+---
+
+# mlx-community/Mamba-Codestral-7B-v0.1-8bits
+
+The Model [mlx-community/Mamba-Codestral-7B-v0.1-8bits](https://huggingface.co/mlx-community/Mamba-Codestral-7B-v0.1-8bits) was converted to MLX format from [mistralai/Mamba-Codestral-7B-v0.1](https://huggingface.co/mistralai/Mamba-Codestral-7B-v0.1) using mlx-lm version **0.18.2**.
+
+## Use with mlx
+
+```bash
+pip install mlx-lm
+```
+
+```python
+from mlx_lm import load, generate
+
+model, tokenizer = load("mlx-community/Mamba-Codestral-7B-v0.1-8bits")
+
+prompt="hello"
+
+if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
+    messages = [{"role": "user", "content": prompt}]
+    prompt = tokenizer.apply_chat_template(
+        messages, tokenize=False, add_generation_prompt=True
+    )
+
+response = generate(model, tokenizer, prompt=prompt, verbose=True)
+```

config.json

@@ -0,0 +1,51 @@
+{
+    "architectures": [
+        "Mamba2ForCausalLM"
+    ],
+    "bos_token_id": 0,
+    "chunk_size": 256,
+    "conv_kernel": 4,
+    "eos_token_id": 0,
+    "expand": 2,
+    "head_dim": 64,
+    "hidden_act": "silu",
+    "hidden_size": 4096,
+    "initializer_range": 0.1,
+    "intermediate_size": 8192,
+    "layer_norm_epsilon": 1e-05,
+    "model_type": "mamba2",
+    "n_groups": 8,
+    "norm_before_gate": true,
+    "num_heads": 128,
+    "num_hidden_layers": 64,
+    "pad_token_id": 0,
+    "quantization": {
+        "group_size": 64,
+        "bits": 8
+    },
+    "quantization_config": {
+        "group_size": 64,
+        "bits": 8
+    },
+    "rescale_prenorm_residual": false,
+    "residual_in_fp32": true,
+    "rms_norm": true,
+    "state_size": 128,
+    "tie_word_embeddings": false,
+    "time_step_floor": 0.0001,
+    "time_step_init_scheme": "random",
+    "time_step_limit": [
+        0.0,
+        Infinity
+    ],
+    "time_step_max": 0.1,
+    "time_step_min": 0.001,
+    "time_step_rank": 256,
+    "time_step_scale": 1.0,
+    "torch_dtype": "bfloat16",
+    "transformers_version": "4.44.0.dev0",
+    "use_bias": false,
+    "use_cache": true,
+    "use_conv_bias": true,
+    "vocab_size": 32768
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:881ad188122c107a87096717ae3aea4e7a1f29391f2a121b3f1fdbaf982ab5ac
+size 5351871176

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83cfaa811eb073d61037d7c1c9dde813233d4ad0aa30b38f76ea951908d53a85
+size 2392798779

model.safetensors.index.json

@@ -0,0 +1,846 @@
+{
+    "metadata": {
+        "total_size": 7744577536
+    },
+    "weight_map": {
+        "backbone.embeddings.biases": "model-00001-of-00002.safetensors",
+        "backbone.embeddings.scales": "model-00001-of-00002.safetensors",
+        "backbone.embeddings.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.0.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.1.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.10.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.11.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.12.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.13.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.14.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.15.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.16.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.17.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.18.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.19.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.2.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.20.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.21.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.22.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.23.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.24.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.25.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.26.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.27.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.28.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.29.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.3.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.30.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.31.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.32.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.33.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.34.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.35.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.36.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.37.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.38.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.39.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.4.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.40.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.41.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.42.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.43.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.44.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.44.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.44.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.44.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.44.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.44.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.44.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.44.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.44.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.44.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.44.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.44.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.44.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.45.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.46.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.47.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.48.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.49.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.5.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.5.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.5.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.5.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.5.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.5.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.5.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.5.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.5.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.5.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.5.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.5.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.5.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.50.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.50.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.50.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.50.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.50.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.50.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.50.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.50.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.50.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.50.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.50.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.50.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.50.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.51.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.52.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.53.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.54.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.55.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.56.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.57.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.58.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.59.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.6.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.6.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.6.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.6.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.6.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.6.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.6.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.6.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.6.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.6.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.6.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.6.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.6.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.60.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.60.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.60.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.60.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.60.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.60.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.60.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.60.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.60.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.60.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.60.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.60.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.60.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.61.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.62.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.mixer.A_log": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.mixer.D": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.mixer.dt_bias": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.mixer.in_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.mixer.in_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.mixer.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.mixer.out_proj.biases": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.mixer.out_proj.scales": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.63.norm.weight": "model-00002-of-00002.safetensors",
+        "backbone.layers.7.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.7.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.7.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.7.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.7.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.7.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.7.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.7.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.7.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.7.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.7.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.7.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.7.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.8.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.mixer.A_log": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.mixer.D": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.mixer.dt_bias": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.mixer.in_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.mixer.in_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.mixer.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.mixer.out_proj.biases": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.mixer.out_proj.scales": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+        "backbone.layers.9.norm.weight": "model-00001-of-00002.safetensors",
+        "backbone.norm_f.weight": "model-00002-of-00002.safetensors",
+        "lm_head.biases": "model-00002-of-00002.safetensors",
+        "lm_head.scales": "model-00002-of-00002.safetensors",
+        "lm_head.weight": "model-00002-of-00002.safetensors"
+    }
+}

special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59f95e28944c062244741268596badc900df86c7f5ded05088d2da22a7379e06
+size 587583