Upload 8 files

config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "sft_results/checkpoint-10000/hf",
+  "architectures": [
+    "MambaForCausalLM"
+  ],
+  "bos_token_id": 0,
+  "conv_kernel": 4,
+  "eos_token_id": 0,
+  "expand": 2,
+  "hidden_act": "silu",
+  "hidden_size": 1536,
+  "initializer_range": 0.1,
+  "intermediate_size": 3072,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "mamba",
+  "num_hidden_layers": 82,
+  "pad_token_id": 0,
+  "rescale_prenorm_residual": false,
+  "residual_in_fp32": true,
+  "state_size": 16,
+  "time_step_floor": 0.0001,
+  "time_step_init_scheme": "random",
+  "time_step_max": 0.1,
+  "time_step_min": 0.001,
+  "time_step_rank": 96,
+  "time_step_scale": 1.0,
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.2",
+  "use_bias": false,
+  "use_cache": true,
+  "use_conv_bias": true,
+  "vocab_size": 32000
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 0,
+  "pad_token_id": 0,
+  "transformers_version": "4.39.2"
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1165abb180cf6960b557e8e7e8000a327c1b226341d57ef35a4b6d6a70571ede
+size 4970124056

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac0cc1a92dc8319d59ccc6b7482bd5e8efe678677d6b68fb21880cebe208045e
+size 119060184

model.safetensors.index.json

@@ -0,0 +1,829 @@
+{
+  "metadata": {
+    "total_size": 5089093632
+  },
+  "weight_map": {
+    "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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.0.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.0.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.0.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.0.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.1.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.1.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.1.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.1.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.10.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.10.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.10.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.10.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.11.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.11.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.11.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.11.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.12.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.12.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.12.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.12.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.13.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.13.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.13.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.13.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.14.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.14.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.14.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.14.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.15.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.15.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.15.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.15.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.16.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.16.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.16.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.16.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.17.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.17.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.17.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.17.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.18.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.18.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.18.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.18.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.19.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.19.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.19.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.19.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.2.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.2.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.2.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.2.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.20.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.20.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.20.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.20.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.21.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.21.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.21.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.21.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.22.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.22.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.22.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.22.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.23.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.23.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.23.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.23.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.24.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.24.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.24.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.24.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.25.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.25.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.25.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.25.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.26.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.26.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.26.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.26.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.27.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.27.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.27.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.27.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.28.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.28.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.28.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.28.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.29.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.29.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.29.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.29.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.3.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.3.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.3.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.3.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.30.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.30.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.30.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.30.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.31.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.31.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.31.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.31.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.32.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.32.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.32.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.32.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.33.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.33.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.33.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.33.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.34.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.34.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.34.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.34.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.35.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.35.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.35.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.35.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.36.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.36.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.36.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.36.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.37.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.37.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.37.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.37.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.38.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.38.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.38.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.38.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.39.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.39.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.39.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.39.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.4.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.4.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.4.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.4.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.40.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.40.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.40.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.40.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.41.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.41.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.41.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.41.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.42.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.42.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.42.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.42.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.43.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.43.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.43.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.43.mixer.x_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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.44.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.44.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.44.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.44.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.44.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.45.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.45.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.45.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.45.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.45.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.45.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.45.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.45.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.45.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.45.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.46.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.46.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.46.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.46.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.46.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.46.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.46.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.46.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.46.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.46.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.47.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.47.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.47.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.47.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.47.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.47.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.47.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.47.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.47.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.47.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.48.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.48.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.48.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.48.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.48.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.48.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.48.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.48.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.48.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.48.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.49.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.49.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.49.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.49.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.49.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.49.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.49.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.49.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.49.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.49.norm.weight": "model-00001-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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.5.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.5.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.5.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.5.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.5.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.50.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.50.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.50.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.50.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.50.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.50.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.50.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.50.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.50.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.50.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.51.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.51.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.51.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.51.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.51.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.51.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.51.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.51.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.51.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.51.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.52.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.52.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.52.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.52.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.52.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.52.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.52.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.52.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.52.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.52.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.53.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.53.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.53.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.53.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.53.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.53.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.53.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.53.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.53.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.53.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.54.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.54.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.54.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.54.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.54.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.54.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.54.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.54.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.54.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.54.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.55.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.55.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.55.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.55.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.55.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.55.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.55.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.55.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.55.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.55.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.56.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.56.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.56.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.56.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.56.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.56.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.56.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.56.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.56.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.56.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.57.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.57.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.57.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.57.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.57.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.57.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.57.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.57.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.57.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.57.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.58.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.58.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.58.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.58.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.58.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.58.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.58.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.58.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.58.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.58.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.59.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.59.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.59.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.59.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.59.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.59.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.59.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.59.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.59.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.59.norm.weight": "model-00001-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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.6.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.6.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.6.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.6.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.6.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.60.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.60.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.60.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.60.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.60.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.60.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.60.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.60.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.60.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.60.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.61.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.61.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.61.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.61.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.61.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.61.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.61.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.61.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.61.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.61.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.62.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.62.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.62.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.62.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.62.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.62.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.62.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.62.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.62.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.62.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.63.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.63.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.63.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.63.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.63.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.63.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.63.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.63.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.63.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.63.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.64.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.64.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.64.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.64.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.64.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.64.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.64.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.64.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.64.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.64.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.65.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.65.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.65.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.65.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.65.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.65.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.65.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.65.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.65.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.65.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.66.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.66.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.66.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.66.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.66.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.66.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.66.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.66.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.66.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.66.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.67.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.67.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.67.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.67.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.67.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.67.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.67.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.67.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.67.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.67.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.68.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.68.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.68.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.68.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.68.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.68.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.68.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.68.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.68.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.68.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.69.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.69.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.69.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.69.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.69.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.69.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.69.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.69.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.69.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.69.norm.weight": "model-00001-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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.7.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.7.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.7.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.7.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.7.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.70.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.70.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.70.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.70.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.70.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.70.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.70.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.70.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.70.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.70.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.71.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.71.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.71.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.71.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.71.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.71.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.71.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.71.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.71.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.71.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.72.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.72.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.72.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.72.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.72.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.72.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.72.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.72.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.72.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.72.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.73.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.73.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.73.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.73.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.73.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.73.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.73.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.73.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.73.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.73.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.74.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.74.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.74.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.74.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.74.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.74.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.74.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.74.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.74.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.74.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.75.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.75.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.75.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.75.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.75.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.75.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.75.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.75.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.75.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.75.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.76.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.76.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.76.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.76.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.76.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.76.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.76.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.76.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.76.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.76.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.77.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.77.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.77.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.77.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.77.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.77.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.77.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.77.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.77.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.77.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.78.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.78.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.78.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.78.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.78.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.78.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.78.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.78.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.78.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.78.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.79.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.79.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.79.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.79.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.79.mixer.dt_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.79.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.79.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.79.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.79.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.79.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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.8.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.8.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.8.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.8.mixer.x_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.8.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.80.mixer.A_log": "model-00001-of-00002.safetensors",
+    "backbone.layers.80.mixer.D": "model-00001-of-00002.safetensors",
+    "backbone.layers.80.mixer.conv1d.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.80.mixer.conv1d.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.80.mixer.dt_proj.bias": "model-00002-of-00002.safetensors",
+    "backbone.layers.80.mixer.dt_proj.weight": "model-00002-of-00002.safetensors",
+    "backbone.layers.80.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+    "backbone.layers.80.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+    "backbone.layers.80.mixer.x_proj.weight": "model-00002-of-00002.safetensors",
+    "backbone.layers.80.norm.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.81.mixer.A_log": "model-00002-of-00002.safetensors",
+    "backbone.layers.81.mixer.D": "model-00002-of-00002.safetensors",
+    "backbone.layers.81.mixer.conv1d.bias": "model-00002-of-00002.safetensors",
+    "backbone.layers.81.mixer.conv1d.weight": "model-00002-of-00002.safetensors",
+    "backbone.layers.81.mixer.dt_proj.bias": "model-00002-of-00002.safetensors",
+    "backbone.layers.81.mixer.dt_proj.weight": "model-00002-of-00002.safetensors",
+    "backbone.layers.81.mixer.in_proj.weight": "model-00002-of-00002.safetensors",
+    "backbone.layers.81.mixer.out_proj.weight": "model-00002-of-00002.safetensors",
+    "backbone.layers.81.mixer.x_proj.weight": "model-00002-of-00002.safetensors",
+    "backbone.layers.81.norm.weight": "model-00002-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_proj.bias": "model-00001-of-00002.safetensors",
+    "backbone.layers.9.mixer.dt_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.9.mixer.in_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.9.mixer.out_proj.weight": "model-00001-of-00002.safetensors",
+    "backbone.layers.9.mixer.x_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"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "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
+  },
+  "pad_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 sha256:2347d2968c37a0e5b5b284719b12d5f0dbc929ccc9928dbff5e7b0ea1ef3e91e
+size 736290

tokenizer_config.json

@@ -0,0 +1,44 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n'  + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}