initial model upload

config.json

@@ -0,0 +1,90 @@
+{
+  "architectures": [
+    "GraniteMoeHybridForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attention_multiplier": 0.0078125,
+  "bos_token_id": 100257,
+  "embedding_multiplier": 12,
+  "eos_token_id": 100257,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "init_method": "mup",
+  "initializer_range": 0.1,
+  "intermediate_size": 768,
+  "layer_types": [
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "attention",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "attention",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "attention",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba",
+    "attention",
+    "mamba",
+    "mamba",
+    "mamba",
+    "mamba"
+  ],
+  "logits_scaling": 16,
+  "mamba_chunk_size": 256,
+  "mamba_conv_bias": true,
+  "mamba_d_conv": 4,
+  "mamba_d_head": 64,
+  "mamba_d_state": 128,
+  "mamba_expand": 2,
+  "mamba_n_groups": 1,
+  "mamba_n_heads": 128,
+  "mamba_proj_bias": false,
+  "max_position_embeddings": 131072,
+  "model_type": "granitemoehybrid",
+  "normalization_function": "rmsnorm",
+  "num_attention_heads": 32,
+  "num_experts_per_tok": 10,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 8,
+  "num_local_experts": 72,
+  "output_router_logits": false,
+  "pad_token_id": 100256,
+  "position_embedding_type": "nope",
+  "residual_multiplier": 0.22,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000,
+  "router_aux_loss_coef": 0.01,
+  "shared_intermediate_size": 1536,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.53.3",
+  "use_cache": true,
+  "vocab_size": 100352
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 100257,
+  "eos_token_id": 100257,
+  "pad_token_id": 100256,
+  "transformers_version": "4.53.3"
+}

model-00001-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22cdb9955764879404d1cbbd744520ec53c7986bf9c3e760533796ed8131ace4
+size 4931823752

model-00002-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e11b406e444df0be6c34eeb2fd4352478a8c512b7e63934b31621d02f611123c
+size 4805655352

model-00003-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c159dabf6acc53763935a03435fd13d6e34fdcd8d5aacdfc87f6f45c0cb94f4
+size 4684967104

model-00004-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f56db2d9910619adece710cce488ab1809aac152274d1b9c469246bc4c013ad9
+size 4805655392

model-00005-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:043bf16b004cb7a837aeea7805736940f6778f7f9eada8ceb0209086c022a1c4
+size 4805655400

model-00006-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9739029ca98cbab6038feb8dcc0b05883c0ba994aaba1d24c4bade08f4b8bece
+size 4684967136

model-00007-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6923892e50ded8070735cec45508a55640039f3e3f01ece3d6f322ce8db55f20
+size 4805655400

model-00008-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15e1a862fb9693f6d00ca07f38192393dd2cb1a413bfdb3de5e1b57efc2dafb3
+size 4805655400

model-00009-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c04b55d66236d1e5995bb16e8df03fdf4fd6f6302ec693136f601a253417744
+size 4684967136

model-00010-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6cd5063fbb77f96a44805f5c78fc5b8cfd80b82d1e9f522d3db4c7703e4b560
+size 4805655400

model-00011-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca406c8f3da58f26b7e243009f1bd1fa85b3c3432890712940335d9c560bd64b
+size 4805655400

model-00012-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d4ef298f04e8b44a2a2467a9d0646bb4dab2605c5f2662e98ef1fa62aaaf6e8
+size 4805655400

model-00013-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d508219e378cdec3ecc92d2cc9c83a2a7688cd4b5b0305e6d38c8add9a93a4ec
+size 4684967144

model-00014-of-00014.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d299279ef4a04d12c39e223cb2a138df0c80cb433fb250912359baf70a89d0a8
+size 2297808752

model.safetensors.index.json

@@ -0,0 +1,594 @@
+{
+  "metadata": {
+    "total_parameters": 32207337984,
+    "total_size": 64414675968
+  },
+  "weight_map": {
+    "model.embed_tokens.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.block_sparse_moe.input_linear.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.block_sparse_moe.output_linear.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.block_sparse_moe.router.layer.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.mamba.A_log": "model-00001-of-00014.safetensors",
+    "model.layers.0.mamba.D": "model-00001-of-00014.safetensors",
+    "model.layers.0.mamba.conv1d.bias": "model-00001-of-00014.safetensors",
+    "model.layers.0.mamba.conv1d.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.mamba.dt_bias": "model-00001-of-00014.safetensors",
+    "model.layers.0.mamba.in_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.mamba.norm.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.mamba.out_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.shared_mlp.input_linear.weight": "model-00001-of-00014.safetensors",
+    "model.layers.0.shared_mlp.output_linear.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.block_sparse_moe.input_linear.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.block_sparse_moe.output_linear.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.block_sparse_moe.router.layer.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.mamba.A_log": "model-00001-of-00014.safetensors",
+    "model.layers.1.mamba.D": "model-00001-of-00014.safetensors",
+    "model.layers.1.mamba.conv1d.bias": "model-00001-of-00014.safetensors",
+    "model.layers.1.mamba.conv1d.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.mamba.dt_bias": "model-00001-of-00014.safetensors",
+    "model.layers.1.mamba.in_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.mamba.norm.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.mamba.out_proj.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.shared_mlp.input_linear.weight": "model-00001-of-00014.safetensors",
+    "model.layers.1.shared_mlp.output_linear.weight": "model-00001-of-00014.safetensors",
+    "model.layers.10.block_sparse_moe.input_linear.weight": "model-00004-of-00014.safetensors",
+    "model.layers.10.block_sparse_moe.output_linear.weight": "model-00004-of-00014.safetensors",
+    "model.layers.10.block_sparse_moe.router.layer.weight": "model-00004-of-00014.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.10.mamba.A_log": "model-00004-of-00014.safetensors",
+    "model.layers.10.mamba.D": "model-00004-of-00014.safetensors",
+    "model.layers.10.mamba.conv1d.bias": "model-00004-of-00014.safetensors",
+    "model.layers.10.mamba.conv1d.weight": "model-00004-of-00014.safetensors",
+    "model.layers.10.mamba.dt_bias": "model-00004-of-00014.safetensors",
+    "model.layers.10.mamba.in_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.10.mamba.norm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.10.mamba.out_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.10.shared_mlp.input_linear.weight": "model-00004-of-00014.safetensors",
+    "model.layers.10.shared_mlp.output_linear.weight": "model-00004-of-00014.safetensors",
+    "model.layers.11.block_sparse_moe.input_linear.weight": "model-00004-of-00014.safetensors",
+    "model.layers.11.block_sparse_moe.output_linear.weight": "model-00005-of-00014.safetensors",
+    "model.layers.11.block_sparse_moe.router.layer.weight": "model-00005-of-00014.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.11.mamba.A_log": "model-00005-of-00014.safetensors",
+    "model.layers.11.mamba.D": "model-00005-of-00014.safetensors",
+    "model.layers.11.mamba.conv1d.bias": "model-00005-of-00014.safetensors",
+    "model.layers.11.mamba.conv1d.weight": "model-00005-of-00014.safetensors",
+    "model.layers.11.mamba.dt_bias": "model-00005-of-00014.safetensors",
+    "model.layers.11.mamba.in_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.11.mamba.norm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.11.mamba.out_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.11.shared_mlp.input_linear.weight": "model-00005-of-00014.safetensors",
+    "model.layers.11.shared_mlp.output_linear.weight": "model-00005-of-00014.safetensors",
+    "model.layers.12.block_sparse_moe.input_linear.weight": "model-00005-of-00014.safetensors",
+    "model.layers.12.block_sparse_moe.output_linear.weight": "model-00005-of-00014.safetensors",
+    "model.layers.12.block_sparse_moe.router.layer.weight": "model-00005-of-00014.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.12.mamba.A_log": "model-00005-of-00014.safetensors",
+    "model.layers.12.mamba.D": "model-00005-of-00014.safetensors",
+    "model.layers.12.mamba.conv1d.bias": "model-00005-of-00014.safetensors",
+    "model.layers.12.mamba.conv1d.weight": "model-00005-of-00014.safetensors",
+    "model.layers.12.mamba.dt_bias": "model-00005-of-00014.safetensors",
+    "model.layers.12.mamba.in_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.12.mamba.norm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.12.mamba.out_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.12.shared_mlp.input_linear.weight": "model-00005-of-00014.safetensors",
+    "model.layers.12.shared_mlp.output_linear.weight": "model-00005-of-00014.safetensors",
+    "model.layers.13.block_sparse_moe.input_linear.weight": "model-00005-of-00014.safetensors",
+    "model.layers.13.block_sparse_moe.output_linear.weight": "model-00005-of-00014.safetensors",
+    "model.layers.13.block_sparse_moe.router.layer.weight": "model-00005-of-00014.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.13.mamba.A_log": "model-00005-of-00014.safetensors",
+    "model.layers.13.mamba.D": "model-00005-of-00014.safetensors",
+    "model.layers.13.mamba.conv1d.bias": "model-00005-of-00014.safetensors",
+    "model.layers.13.mamba.conv1d.weight": "model-00005-of-00014.safetensors",
+    "model.layers.13.mamba.dt_bias": "model-00005-of-00014.safetensors",
+    "model.layers.13.mamba.in_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.13.mamba.norm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.13.mamba.out_proj.weight": "model-00005-of-00014.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00005-of-00014.safetensors",
+    "model.layers.13.shared_mlp.input_linear.weight": "model-00005-of-00014.safetensors",
+    "model.layers.13.shared_mlp.output_linear.weight": "model-00005-of-00014.safetensors",
+    "model.layers.14.block_sparse_moe.input_linear.weight": "model-00005-of-00014.safetensors",
+    "model.layers.14.block_sparse_moe.output_linear.weight": "model-00006-of-00014.safetensors",
+    "model.layers.14.block_sparse_moe.router.layer.weight": "model-00006-of-00014.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.14.mamba.A_log": "model-00006-of-00014.safetensors",
+    "model.layers.14.mamba.D": "model-00006-of-00014.safetensors",
+    "model.layers.14.mamba.conv1d.bias": "model-00006-of-00014.safetensors",
+    "model.layers.14.mamba.conv1d.weight": "model-00006-of-00014.safetensors",
+    "model.layers.14.mamba.dt_bias": "model-00006-of-00014.safetensors",
+    "model.layers.14.mamba.in_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.14.mamba.norm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.14.mamba.out_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.14.shared_mlp.input_linear.weight": "model-00006-of-00014.safetensors",
+    "model.layers.14.shared_mlp.output_linear.weight": "model-00006-of-00014.safetensors",
+    "model.layers.15.block_sparse_moe.input_linear.weight": "model-00006-of-00014.safetensors",
+    "model.layers.15.block_sparse_moe.output_linear.weight": "model-00006-of-00014.safetensors",
+    "model.layers.15.block_sparse_moe.router.layer.weight": "model-00006-of-00014.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.15.shared_mlp.input_linear.weight": "model-00006-of-00014.safetensors",
+    "model.layers.15.shared_mlp.output_linear.weight": "model-00006-of-00014.safetensors",
+    "model.layers.16.block_sparse_moe.input_linear.weight": "model-00006-of-00014.safetensors",
+    "model.layers.16.block_sparse_moe.output_linear.weight": "model-00006-of-00014.safetensors",
+    "model.layers.16.block_sparse_moe.router.layer.weight": "model-00006-of-00014.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.16.mamba.A_log": "model-00006-of-00014.safetensors",
+    "model.layers.16.mamba.D": "model-00006-of-00014.safetensors",
+    "model.layers.16.mamba.conv1d.bias": "model-00006-of-00014.safetensors",
+    "model.layers.16.mamba.conv1d.weight": "model-00006-of-00014.safetensors",
+    "model.layers.16.mamba.dt_bias": "model-00006-of-00014.safetensors",
+    "model.layers.16.mamba.in_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.16.mamba.norm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.16.mamba.out_proj.weight": "model-00006-of-00014.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00006-of-00014.safetensors",
+    "model.layers.16.shared_mlp.input_linear.weight": "model-00006-of-00014.safetensors",
+    "model.layers.16.shared_mlp.output_linear.weight": "model-00006-of-00014.safetensors",
+    "model.layers.17.block_sparse_moe.input_linear.weight": "model-00006-of-00014.safetensors",
+    "model.layers.17.block_sparse_moe.output_linear.weight": "model-00007-of-00014.safetensors",
+    "model.layers.17.block_sparse_moe.router.layer.weight": "model-00007-of-00014.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.17.mamba.A_log": "model-00007-of-00014.safetensors",
+    "model.layers.17.mamba.D": "model-00007-of-00014.safetensors",
+    "model.layers.17.mamba.conv1d.bias": "model-00007-of-00014.safetensors",
+    "model.layers.17.mamba.conv1d.weight": "model-00007-of-00014.safetensors",
+    "model.layers.17.mamba.dt_bias": "model-00007-of-00014.safetensors",
+    "model.layers.17.mamba.in_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.17.mamba.norm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.17.mamba.out_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.17.shared_mlp.input_linear.weight": "model-00007-of-00014.safetensors",
+    "model.layers.17.shared_mlp.output_linear.weight": "model-00007-of-00014.safetensors",
+    "model.layers.18.block_sparse_moe.input_linear.weight": "model-00007-of-00014.safetensors",
+    "model.layers.18.block_sparse_moe.output_linear.weight": "model-00007-of-00014.safetensors",
+    "model.layers.18.block_sparse_moe.router.layer.weight": "model-00007-of-00014.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.18.mamba.A_log": "model-00007-of-00014.safetensors",
+    "model.layers.18.mamba.D": "model-00007-of-00014.safetensors",
+    "model.layers.18.mamba.conv1d.bias": "model-00007-of-00014.safetensors",
+    "model.layers.18.mamba.conv1d.weight": "model-00007-of-00014.safetensors",
+    "model.layers.18.mamba.dt_bias": "model-00007-of-00014.safetensors",
+    "model.layers.18.mamba.in_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.18.mamba.norm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.18.mamba.out_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.18.shared_mlp.input_linear.weight": "model-00007-of-00014.safetensors",
+    "model.layers.18.shared_mlp.output_linear.weight": "model-00007-of-00014.safetensors",
+    "model.layers.19.block_sparse_moe.input_linear.weight": "model-00007-of-00014.safetensors",
+    "model.layers.19.block_sparse_moe.output_linear.weight": "model-00007-of-00014.safetensors",
+    "model.layers.19.block_sparse_moe.router.layer.weight": "model-00007-of-00014.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.19.mamba.A_log": "model-00007-of-00014.safetensors",
+    "model.layers.19.mamba.D": "model-00007-of-00014.safetensors",
+    "model.layers.19.mamba.conv1d.bias": "model-00007-of-00014.safetensors",
+    "model.layers.19.mamba.conv1d.weight": "model-00007-of-00014.safetensors",
+    "model.layers.19.mamba.dt_bias": "model-00007-of-00014.safetensors",
+    "model.layers.19.mamba.in_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.19.mamba.norm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.19.mamba.out_proj.weight": "model-00007-of-00014.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00007-of-00014.safetensors",
+    "model.layers.19.shared_mlp.input_linear.weight": "model-00007-of-00014.safetensors",
+    "model.layers.19.shared_mlp.output_linear.weight": "model-00007-of-00014.safetensors",
+    "model.layers.2.block_sparse_moe.input_linear.weight": "model-00001-of-00014.safetensors",
+    "model.layers.2.block_sparse_moe.output_linear.weight": "model-00002-of-00014.safetensors",
+    "model.layers.2.block_sparse_moe.router.layer.weight": "model-00002-of-00014.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.2.mamba.A_log": "model-00002-of-00014.safetensors",
+    "model.layers.2.mamba.D": "model-00002-of-00014.safetensors",
+    "model.layers.2.mamba.conv1d.bias": "model-00002-of-00014.safetensors",
+    "model.layers.2.mamba.conv1d.weight": "model-00002-of-00014.safetensors",
+    "model.layers.2.mamba.dt_bias": "model-00002-of-00014.safetensors",
+    "model.layers.2.mamba.in_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.2.mamba.norm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.2.mamba.out_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.2.shared_mlp.input_linear.weight": "model-00002-of-00014.safetensors",
+    "model.layers.2.shared_mlp.output_linear.weight": "model-00002-of-00014.safetensors",
+    "model.layers.20.block_sparse_moe.input_linear.weight": "model-00007-of-00014.safetensors",
+    "model.layers.20.block_sparse_moe.output_linear.weight": "model-00008-of-00014.safetensors",
+    "model.layers.20.block_sparse_moe.router.layer.weight": "model-00008-of-00014.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.20.mamba.A_log": "model-00008-of-00014.safetensors",
+    "model.layers.20.mamba.D": "model-00008-of-00014.safetensors",
+    "model.layers.20.mamba.conv1d.bias": "model-00008-of-00014.safetensors",
+    "model.layers.20.mamba.conv1d.weight": "model-00008-of-00014.safetensors",
+    "model.layers.20.mamba.dt_bias": "model-00008-of-00014.safetensors",
+    "model.layers.20.mamba.in_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.20.mamba.norm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.20.mamba.out_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.20.shared_mlp.input_linear.weight": "model-00008-of-00014.safetensors",
+    "model.layers.20.shared_mlp.output_linear.weight": "model-00008-of-00014.safetensors",
+    "model.layers.21.block_sparse_moe.input_linear.weight": "model-00008-of-00014.safetensors",
+    "model.layers.21.block_sparse_moe.output_linear.weight": "model-00008-of-00014.safetensors",
+    "model.layers.21.block_sparse_moe.router.layer.weight": "model-00008-of-00014.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.21.mamba.A_log": "model-00008-of-00014.safetensors",
+    "model.layers.21.mamba.D": "model-00008-of-00014.safetensors",
+    "model.layers.21.mamba.conv1d.bias": "model-00008-of-00014.safetensors",
+    "model.layers.21.mamba.conv1d.weight": "model-00008-of-00014.safetensors",
+    "model.layers.21.mamba.dt_bias": "model-00008-of-00014.safetensors",
+    "model.layers.21.mamba.in_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.21.mamba.norm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.21.mamba.out_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.21.shared_mlp.input_linear.weight": "model-00008-of-00014.safetensors",
+    "model.layers.21.shared_mlp.output_linear.weight": "model-00008-of-00014.safetensors",
+    "model.layers.22.block_sparse_moe.input_linear.weight": "model-00008-of-00014.safetensors",
+    "model.layers.22.block_sparse_moe.output_linear.weight": "model-00008-of-00014.safetensors",
+    "model.layers.22.block_sparse_moe.router.layer.weight": "model-00008-of-00014.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.22.mamba.A_log": "model-00008-of-00014.safetensors",
+    "model.layers.22.mamba.D": "model-00008-of-00014.safetensors",
+    "model.layers.22.mamba.conv1d.bias": "model-00008-of-00014.safetensors",
+    "model.layers.22.mamba.conv1d.weight": "model-00008-of-00014.safetensors",
+    "model.layers.22.mamba.dt_bias": "model-00008-of-00014.safetensors",
+    "model.layers.22.mamba.in_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.22.mamba.norm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.22.mamba.out_proj.weight": "model-00008-of-00014.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00008-of-00014.safetensors",
+    "model.layers.22.shared_mlp.input_linear.weight": "model-00008-of-00014.safetensors",
+    "model.layers.22.shared_mlp.output_linear.weight": "model-00008-of-00014.safetensors",
+    "model.layers.23.block_sparse_moe.input_linear.weight": "model-00008-of-00014.safetensors",
+    "model.layers.23.block_sparse_moe.output_linear.weight": "model-00009-of-00014.safetensors",
+    "model.layers.23.block_sparse_moe.router.layer.weight": "model-00009-of-00014.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.23.mamba.A_log": "model-00009-of-00014.safetensors",
+    "model.layers.23.mamba.D": "model-00009-of-00014.safetensors",
+    "model.layers.23.mamba.conv1d.bias": "model-00009-of-00014.safetensors",
+    "model.layers.23.mamba.conv1d.weight": "model-00009-of-00014.safetensors",
+    "model.layers.23.mamba.dt_bias": "model-00009-of-00014.safetensors",
+    "model.layers.23.mamba.in_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.23.mamba.norm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.23.mamba.out_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.23.shared_mlp.input_linear.weight": "model-00009-of-00014.safetensors",
+    "model.layers.23.shared_mlp.output_linear.weight": "model-00009-of-00014.safetensors",
+    "model.layers.24.block_sparse_moe.input_linear.weight": "model-00009-of-00014.safetensors",
+    "model.layers.24.block_sparse_moe.output_linear.weight": "model-00009-of-00014.safetensors",
+    "model.layers.24.block_sparse_moe.router.layer.weight": "model-00009-of-00014.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.24.mamba.A_log": "model-00009-of-00014.safetensors",
+    "model.layers.24.mamba.D": "model-00009-of-00014.safetensors",
+    "model.layers.24.mamba.conv1d.bias": "model-00009-of-00014.safetensors",
+    "model.layers.24.mamba.conv1d.weight": "model-00009-of-00014.safetensors",
+    "model.layers.24.mamba.dt_bias": "model-00009-of-00014.safetensors",
+    "model.layers.24.mamba.in_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.24.mamba.norm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.24.mamba.out_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.24.shared_mlp.input_linear.weight": "model-00009-of-00014.safetensors",
+    "model.layers.24.shared_mlp.output_linear.weight": "model-00009-of-00014.safetensors",
+    "model.layers.25.block_sparse_moe.input_linear.weight": "model-00009-of-00014.safetensors",
+    "model.layers.25.block_sparse_moe.output_linear.weight": "model-00009-of-00014.safetensors",
+    "model.layers.25.block_sparse_moe.router.layer.weight": "model-00009-of-00014.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00009-of-00014.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00009-of-00014.safetensors",
+    "model.layers.25.shared_mlp.input_linear.weight": "model-00009-of-00014.safetensors",
+    "model.layers.25.shared_mlp.output_linear.weight": "model-00009-of-00014.safetensors",
+    "model.layers.26.block_sparse_moe.input_linear.weight": "model-00009-of-00014.safetensors",
+    "model.layers.26.block_sparse_moe.output_linear.weight": "model-00010-of-00014.safetensors",
+    "model.layers.26.block_sparse_moe.router.layer.weight": "model-00010-of-00014.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.26.mamba.A_log": "model-00010-of-00014.safetensors",
+    "model.layers.26.mamba.D": "model-00010-of-00014.safetensors",
+    "model.layers.26.mamba.conv1d.bias": "model-00010-of-00014.safetensors",
+    "model.layers.26.mamba.conv1d.weight": "model-00010-of-00014.safetensors",
+    "model.layers.26.mamba.dt_bias": "model-00010-of-00014.safetensors",
+    "model.layers.26.mamba.in_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.26.mamba.norm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.26.mamba.out_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.26.shared_mlp.input_linear.weight": "model-00010-of-00014.safetensors",
+    "model.layers.26.shared_mlp.output_linear.weight": "model-00010-of-00014.safetensors",
+    "model.layers.27.block_sparse_moe.input_linear.weight": "model-00010-of-00014.safetensors",
+    "model.layers.27.block_sparse_moe.output_linear.weight": "model-00010-of-00014.safetensors",
+    "model.layers.27.block_sparse_moe.router.layer.weight": "model-00010-of-00014.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.27.mamba.A_log": "model-00010-of-00014.safetensors",
+    "model.layers.27.mamba.D": "model-00010-of-00014.safetensors",
+    "model.layers.27.mamba.conv1d.bias": "model-00010-of-00014.safetensors",
+    "model.layers.27.mamba.conv1d.weight": "model-00010-of-00014.safetensors",
+    "model.layers.27.mamba.dt_bias": "model-00010-of-00014.safetensors",
+    "model.layers.27.mamba.in_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.27.mamba.norm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.27.mamba.out_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.27.shared_mlp.input_linear.weight": "model-00010-of-00014.safetensors",
+    "model.layers.27.shared_mlp.output_linear.weight": "model-00010-of-00014.safetensors",
+    "model.layers.28.block_sparse_moe.input_linear.weight": "model-00010-of-00014.safetensors",
+    "model.layers.28.block_sparse_moe.output_linear.weight": "model-00010-of-00014.safetensors",
+    "model.layers.28.block_sparse_moe.router.layer.weight": "model-00010-of-00014.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.28.mamba.A_log": "model-00010-of-00014.safetensors",
+    "model.layers.28.mamba.D": "model-00010-of-00014.safetensors",
+    "model.layers.28.mamba.conv1d.bias": "model-00010-of-00014.safetensors",
+    "model.layers.28.mamba.conv1d.weight": "model-00010-of-00014.safetensors",
+    "model.layers.28.mamba.dt_bias": "model-00010-of-00014.safetensors",
+    "model.layers.28.mamba.in_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.28.mamba.norm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.28.mamba.out_proj.weight": "model-00010-of-00014.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00010-of-00014.safetensors",
+    "model.layers.28.shared_mlp.input_linear.weight": "model-00010-of-00014.safetensors",
+    "model.layers.28.shared_mlp.output_linear.weight": "model-00010-of-00014.safetensors",
+    "model.layers.29.block_sparse_moe.input_linear.weight": "model-00010-of-00014.safetensors",
+    "model.layers.29.block_sparse_moe.output_linear.weight": "model-00011-of-00014.safetensors",
+    "model.layers.29.block_sparse_moe.router.layer.weight": "model-00011-of-00014.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.29.mamba.A_log": "model-00011-of-00014.safetensors",
+    "model.layers.29.mamba.D": "model-00011-of-00014.safetensors",
+    "model.layers.29.mamba.conv1d.bias": "model-00011-of-00014.safetensors",
+    "model.layers.29.mamba.conv1d.weight": "model-00011-of-00014.safetensors",
+    "model.layers.29.mamba.dt_bias": "model-00011-of-00014.safetensors",
+    "model.layers.29.mamba.in_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.29.mamba.norm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.29.mamba.out_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.29.shared_mlp.input_linear.weight": "model-00011-of-00014.safetensors",
+    "model.layers.29.shared_mlp.output_linear.weight": "model-00011-of-00014.safetensors",
+    "model.layers.3.block_sparse_moe.input_linear.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.block_sparse_moe.output_linear.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.block_sparse_moe.router.layer.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.mamba.A_log": "model-00002-of-00014.safetensors",
+    "model.layers.3.mamba.D": "model-00002-of-00014.safetensors",
+    "model.layers.3.mamba.conv1d.bias": "model-00002-of-00014.safetensors",
+    "model.layers.3.mamba.conv1d.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.mamba.dt_bias": "model-00002-of-00014.safetensors",
+    "model.layers.3.mamba.in_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.mamba.norm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.mamba.out_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.shared_mlp.input_linear.weight": "model-00002-of-00014.safetensors",
+    "model.layers.3.shared_mlp.output_linear.weight": "model-00002-of-00014.safetensors",
+    "model.layers.30.block_sparse_moe.input_linear.weight": "model-00011-of-00014.safetensors",
+    "model.layers.30.block_sparse_moe.output_linear.weight": "model-00011-of-00014.safetensors",
+    "model.layers.30.block_sparse_moe.router.layer.weight": "model-00011-of-00014.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.30.mamba.A_log": "model-00011-of-00014.safetensors",
+    "model.layers.30.mamba.D": "model-00011-of-00014.safetensors",
+    "model.layers.30.mamba.conv1d.bias": "model-00011-of-00014.safetensors",
+    "model.layers.30.mamba.conv1d.weight": "model-00011-of-00014.safetensors",
+    "model.layers.30.mamba.dt_bias": "model-00011-of-00014.safetensors",
+    "model.layers.30.mamba.in_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.30.mamba.norm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.30.mamba.out_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.30.shared_mlp.input_linear.weight": "model-00011-of-00014.safetensors",
+    "model.layers.30.shared_mlp.output_linear.weight": "model-00011-of-00014.safetensors",
+    "model.layers.31.block_sparse_moe.input_linear.weight": "model-00011-of-00014.safetensors",
+    "model.layers.31.block_sparse_moe.output_linear.weight": "model-00011-of-00014.safetensors",
+    "model.layers.31.block_sparse_moe.router.layer.weight": "model-00011-of-00014.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.31.mamba.A_log": "model-00011-of-00014.safetensors",
+    "model.layers.31.mamba.D": "model-00011-of-00014.safetensors",
+    "model.layers.31.mamba.conv1d.bias": "model-00011-of-00014.safetensors",
+    "model.layers.31.mamba.conv1d.weight": "model-00011-of-00014.safetensors",
+    "model.layers.31.mamba.dt_bias": "model-00011-of-00014.safetensors",
+    "model.layers.31.mamba.in_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.31.mamba.norm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.31.mamba.out_proj.weight": "model-00011-of-00014.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00011-of-00014.safetensors",
+    "model.layers.31.shared_mlp.input_linear.weight": "model-00011-of-00014.safetensors",
+    "model.layers.31.shared_mlp.output_linear.weight": "model-00011-of-00014.safetensors",
+    "model.layers.32.block_sparse_moe.input_linear.weight": "model-00011-of-00014.safetensors",
+    "model.layers.32.block_sparse_moe.output_linear.weight": "model-00012-of-00014.safetensors",
+    "model.layers.32.block_sparse_moe.router.layer.weight": "model-00012-of-00014.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.32.mamba.A_log": "model-00012-of-00014.safetensors",
+    "model.layers.32.mamba.D": "model-00012-of-00014.safetensors",
+    "model.layers.32.mamba.conv1d.bias": "model-00012-of-00014.safetensors",
+    "model.layers.32.mamba.conv1d.weight": "model-00012-of-00014.safetensors",
+    "model.layers.32.mamba.dt_bias": "model-00012-of-00014.safetensors",
+    "model.layers.32.mamba.in_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.32.mamba.norm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.32.mamba.out_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.32.shared_mlp.input_linear.weight": "model-00012-of-00014.safetensors",
+    "model.layers.32.shared_mlp.output_linear.weight": "model-00012-of-00014.safetensors",
+    "model.layers.33.block_sparse_moe.input_linear.weight": "model-00012-of-00014.safetensors",
+    "model.layers.33.block_sparse_moe.output_linear.weight": "model-00012-of-00014.safetensors",
+    "model.layers.33.block_sparse_moe.router.layer.weight": "model-00012-of-00014.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.33.mamba.A_log": "model-00012-of-00014.safetensors",
+    "model.layers.33.mamba.D": "model-00012-of-00014.safetensors",
+    "model.layers.33.mamba.conv1d.bias": "model-00012-of-00014.safetensors",
+    "model.layers.33.mamba.conv1d.weight": "model-00012-of-00014.safetensors",
+    "model.layers.33.mamba.dt_bias": "model-00012-of-00014.safetensors",
+    "model.layers.33.mamba.in_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.33.mamba.norm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.33.mamba.out_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.33.shared_mlp.input_linear.weight": "model-00012-of-00014.safetensors",
+    "model.layers.33.shared_mlp.output_linear.weight": "model-00012-of-00014.safetensors",
+    "model.layers.34.block_sparse_moe.input_linear.weight": "model-00012-of-00014.safetensors",
+    "model.layers.34.block_sparse_moe.output_linear.weight": "model-00012-of-00014.safetensors",
+    "model.layers.34.block_sparse_moe.router.layer.weight": "model-00012-of-00014.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.34.mamba.A_log": "model-00012-of-00014.safetensors",
+    "model.layers.34.mamba.D": "model-00012-of-00014.safetensors",
+    "model.layers.34.mamba.conv1d.bias": "model-00012-of-00014.safetensors",
+    "model.layers.34.mamba.conv1d.weight": "model-00012-of-00014.safetensors",
+    "model.layers.34.mamba.dt_bias": "model-00012-of-00014.safetensors",
+    "model.layers.34.mamba.in_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.34.mamba.norm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.34.mamba.out_proj.weight": "model-00012-of-00014.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00012-of-00014.safetensors",
+    "model.layers.34.shared_mlp.input_linear.weight": "model-00012-of-00014.safetensors",
+    "model.layers.34.shared_mlp.output_linear.weight": "model-00012-of-00014.safetensors",
+    "model.layers.35.block_sparse_moe.input_linear.weight": "model-00012-of-00014.safetensors",
+    "model.layers.35.block_sparse_moe.output_linear.weight": "model-00013-of-00014.safetensors",
+    "model.layers.35.block_sparse_moe.router.layer.weight": "model-00013-of-00014.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.35.shared_mlp.input_linear.weight": "model-00013-of-00014.safetensors",
+    "model.layers.35.shared_mlp.output_linear.weight": "model-00013-of-00014.safetensors",
+    "model.layers.36.block_sparse_moe.input_linear.weight": "model-00013-of-00014.safetensors",
+    "model.layers.36.block_sparse_moe.output_linear.weight": "model-00013-of-00014.safetensors",
+    "model.layers.36.block_sparse_moe.router.layer.weight": "model-00013-of-00014.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.36.mamba.A_log": "model-00013-of-00014.safetensors",
+    "model.layers.36.mamba.D": "model-00013-of-00014.safetensors",
+    "model.layers.36.mamba.conv1d.bias": "model-00013-of-00014.safetensors",
+    "model.layers.36.mamba.conv1d.weight": "model-00013-of-00014.safetensors",
+    "model.layers.36.mamba.dt_bias": "model-00013-of-00014.safetensors",
+    "model.layers.36.mamba.in_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.36.mamba.norm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.36.mamba.out_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.36.shared_mlp.input_linear.weight": "model-00013-of-00014.safetensors",
+    "model.layers.36.shared_mlp.output_linear.weight": "model-00013-of-00014.safetensors",
+    "model.layers.37.block_sparse_moe.input_linear.weight": "model-00013-of-00014.safetensors",
+    "model.layers.37.block_sparse_moe.output_linear.weight": "model-00013-of-00014.safetensors",
+    "model.layers.37.block_sparse_moe.router.layer.weight": "model-00013-of-00014.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.37.mamba.A_log": "model-00013-of-00014.safetensors",
+    "model.layers.37.mamba.D": "model-00013-of-00014.safetensors",
+    "model.layers.37.mamba.conv1d.bias": "model-00013-of-00014.safetensors",
+    "model.layers.37.mamba.conv1d.weight": "model-00013-of-00014.safetensors",
+    "model.layers.37.mamba.dt_bias": "model-00013-of-00014.safetensors",
+    "model.layers.37.mamba.in_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.37.mamba.norm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.37.mamba.out_proj.weight": "model-00013-of-00014.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00013-of-00014.safetensors",
+    "model.layers.37.shared_mlp.input_linear.weight": "model-00013-of-00014.safetensors",
+    "model.layers.37.shared_mlp.output_linear.weight": "model-00013-of-00014.safetensors",
+    "model.layers.38.block_sparse_moe.input_linear.weight": "model-00013-of-00014.safetensors",
+    "model.layers.38.block_sparse_moe.output_linear.weight": "model-00014-of-00014.safetensors",
+    "model.layers.38.block_sparse_moe.router.layer.weight": "model-00014-of-00014.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00014-of-00014.safetensors",
+    "model.layers.38.mamba.A_log": "model-00014-of-00014.safetensors",
+    "model.layers.38.mamba.D": "model-00014-of-00014.safetensors",
+    "model.layers.38.mamba.conv1d.bias": "model-00014-of-00014.safetensors",
+    "model.layers.38.mamba.conv1d.weight": "model-00014-of-00014.safetensors",
+    "model.layers.38.mamba.dt_bias": "model-00014-of-00014.safetensors",
+    "model.layers.38.mamba.in_proj.weight": "model-00014-of-00014.safetensors",
+    "model.layers.38.mamba.norm.weight": "model-00014-of-00014.safetensors",
+    "model.layers.38.mamba.out_proj.weight": "model-00014-of-00014.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00014-of-00014.safetensors",
+    "model.layers.38.shared_mlp.input_linear.weight": "model-00014-of-00014.safetensors",
+    "model.layers.38.shared_mlp.output_linear.weight": "model-00014-of-00014.safetensors",
+    "model.layers.39.block_sparse_moe.input_linear.weight": "model-00014-of-00014.safetensors",
+    "model.layers.39.block_sparse_moe.output_linear.weight": "model-00014-of-00014.safetensors",
+    "model.layers.39.block_sparse_moe.router.layer.weight": "model-00014-of-00014.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00014-of-00014.safetensors",
+    "model.layers.39.mamba.A_log": "model-00014-of-00014.safetensors",
+    "model.layers.39.mamba.D": "model-00014-of-00014.safetensors",
+    "model.layers.39.mamba.conv1d.bias": "model-00014-of-00014.safetensors",
+    "model.layers.39.mamba.conv1d.weight": "model-00014-of-00014.safetensors",
+    "model.layers.39.mamba.dt_bias": "model-00014-of-00014.safetensors",
+    "model.layers.39.mamba.in_proj.weight": "model-00014-of-00014.safetensors",
+    "model.layers.39.mamba.norm.weight": "model-00014-of-00014.safetensors",
+    "model.layers.39.mamba.out_proj.weight": "model-00014-of-00014.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00014-of-00014.safetensors",
+    "model.layers.39.shared_mlp.input_linear.weight": "model-00014-of-00014.safetensors",
+    "model.layers.39.shared_mlp.output_linear.weight": "model-00014-of-00014.safetensors",
+    "model.layers.4.block_sparse_moe.input_linear.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.block_sparse_moe.output_linear.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.block_sparse_moe.router.layer.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.mamba.A_log": "model-00002-of-00014.safetensors",
+    "model.layers.4.mamba.D": "model-00002-of-00014.safetensors",
+    "model.layers.4.mamba.conv1d.bias": "model-00002-of-00014.safetensors",
+    "model.layers.4.mamba.conv1d.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.mamba.dt_bias": "model-00002-of-00014.safetensors",
+    "model.layers.4.mamba.in_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.mamba.norm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.mamba.out_proj.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.shared_mlp.input_linear.weight": "model-00002-of-00014.safetensors",
+    "model.layers.4.shared_mlp.output_linear.weight": "model-00002-of-00014.safetensors",
+    "model.layers.5.block_sparse_moe.input_linear.weight": "model-00002-of-00014.safetensors",
+    "model.layers.5.block_sparse_moe.output_linear.weight": "model-00003-of-00014.safetensors",
+    "model.layers.5.block_sparse_moe.router.layer.weight": "model-00003-of-00014.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.5.shared_mlp.input_linear.weight": "model-00003-of-00014.safetensors",
+    "model.layers.5.shared_mlp.output_linear.weight": "model-00003-of-00014.safetensors",
+    "model.layers.6.block_sparse_moe.input_linear.weight": "model-00003-of-00014.safetensors",
+    "model.layers.6.block_sparse_moe.output_linear.weight": "model-00003-of-00014.safetensors",
+    "model.layers.6.block_sparse_moe.router.layer.weight": "model-00003-of-00014.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.6.mamba.A_log": "model-00003-of-00014.safetensors",
+    "model.layers.6.mamba.D": "model-00003-of-00014.safetensors",
+    "model.layers.6.mamba.conv1d.bias": "model-00003-of-00014.safetensors",
+    "model.layers.6.mamba.conv1d.weight": "model-00003-of-00014.safetensors",
+    "model.layers.6.mamba.dt_bias": "model-00003-of-00014.safetensors",
+    "model.layers.6.mamba.in_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.6.mamba.norm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.6.mamba.out_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.6.shared_mlp.input_linear.weight": "model-00003-of-00014.safetensors",
+    "model.layers.6.shared_mlp.output_linear.weight": "model-00003-of-00014.safetensors",
+    "model.layers.7.block_sparse_moe.input_linear.weight": "model-00003-of-00014.safetensors",
+    "model.layers.7.block_sparse_moe.output_linear.weight": "model-00003-of-00014.safetensors",
+    "model.layers.7.block_sparse_moe.router.layer.weight": "model-00003-of-00014.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.7.mamba.A_log": "model-00003-of-00014.safetensors",
+    "model.layers.7.mamba.D": "model-00003-of-00014.safetensors",
+    "model.layers.7.mamba.conv1d.bias": "model-00003-of-00014.safetensors",
+    "model.layers.7.mamba.conv1d.weight": "model-00003-of-00014.safetensors",
+    "model.layers.7.mamba.dt_bias": "model-00003-of-00014.safetensors",
+    "model.layers.7.mamba.in_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.7.mamba.norm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.7.mamba.out_proj.weight": "model-00003-of-00014.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00003-of-00014.safetensors",
+    "model.layers.7.shared_mlp.input_linear.weight": "model-00003-of-00014.safetensors",
+    "model.layers.7.shared_mlp.output_linear.weight": "model-00003-of-00014.safetensors",
+    "model.layers.8.block_sparse_moe.input_linear.weight": "model-00003-of-00014.safetensors",
+    "model.layers.8.block_sparse_moe.output_linear.weight": "model-00004-of-00014.safetensors",
+    "model.layers.8.block_sparse_moe.router.layer.weight": "model-00004-of-00014.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.8.mamba.A_log": "model-00004-of-00014.safetensors",
+    "model.layers.8.mamba.D": "model-00004-of-00014.safetensors",
+    "model.layers.8.mamba.conv1d.bias": "model-00004-of-00014.safetensors",
+    "model.layers.8.mamba.conv1d.weight": "model-00004-of-00014.safetensors",
+    "model.layers.8.mamba.dt_bias": "model-00004-of-00014.safetensors",
+    "model.layers.8.mamba.in_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.8.mamba.norm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.8.mamba.out_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.8.shared_mlp.input_linear.weight": "model-00004-of-00014.safetensors",
+    "model.layers.8.shared_mlp.output_linear.weight": "model-00004-of-00014.safetensors",
+    "model.layers.9.block_sparse_moe.input_linear.weight": "model-00004-of-00014.safetensors",
+    "model.layers.9.block_sparse_moe.output_linear.weight": "model-00004-of-00014.safetensors",
+    "model.layers.9.block_sparse_moe.router.layer.weight": "model-00004-of-00014.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.9.mamba.A_log": "model-00004-of-00014.safetensors",
+    "model.layers.9.mamba.D": "model-00004-of-00014.safetensors",
+    "model.layers.9.mamba.conv1d.bias": "model-00004-of-00014.safetensors",
+    "model.layers.9.mamba.conv1d.weight": "model-00004-of-00014.safetensors",
+    "model.layers.9.mamba.dt_bias": "model-00004-of-00014.safetensors",
+    "model.layers.9.mamba.in_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.9.mamba.norm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.9.mamba.out_proj.weight": "model-00004-of-00014.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00004-of-00014.safetensors",
+    "model.layers.9.shared_mlp.input_linear.weight": "model-00004-of-00014.safetensors",
+    "model.layers.9.shared_mlp.output_linear.weight": "model-00004-of-00014.safetensors",
+    "model.norm.weight": "model-00014-of-00014.safetensors"
+  }
+}

model.sig

@@ -0,0 +1 @@
+{"mediaType": "application/vnd.dev.sigstore.bundle.v0.3+json", "verificationMaterial": {"certificate": {"rawBytes": "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"}, "tlogEntries": [{"logIndex": "541452800", "logId": {"keyId": "wNI9atQGlz+VWfO6LRygH4QUfY/8W4RFwiT5i5WRgB0="}, "kindVersion": {"kind": "dsse", "version": "0.0.1"}, "integratedTime": "1758390967", "inclusionPromise": {"signedEntryTimestamp": "MEQCIBvC12hWF8TviFIukAGZNbMCD3Tx/tf6B1ER1nBmsQmkAiBaUNZZtVBLyOUshBneejH8u7U8bT0cFaD6fFUax2DD8g=="}, "inclusionProof": {"logIndex": "419548538", "rootHash": "wlKXR54PophNQ/0UdItOEjggE4Nmv8UkqyyDMY4pgnM=", "treeSize": "419548542", "hashes": ["VcuY4YGz+KkkT4Ccx7q/BEl6oxfTHIqvrwbI/K2s2FY=", "j0JNU7rhktODpLk+vont42oa0wvrIkWw5Kay+SrXKGc=", "INKqHEp174nKSvCiseWXVoDBKrQ6jarI76ByNoNCaMI=", "0FZEs6wjS01FsYomBXDxvTPcOxq0xwiH9rnB4p2RU7Q=", "KLnGDMytH125BPEw6Fn+oFLord0AOBw67iPM4VyGBCQ=", "yTGpkKtsc1k6nxMqmzgOUnjOKVU7nBS7HUTor/VFqB8=", "p/KPnKxw2etfvLjISZ9cr7sTalQDSqwoc/q3WOiIPLU=", "7dR/rKibaeMoFV5/9gEqsCBJNQ62RkO1BwomAMgX7Mo=", "XnSkMXruXo+espDCVwTNIbxP5YGvL2EwtekWe7fQXXI=", "1dxXAViDba+WHBHwrLg823Mw+kRQ6CwBWab+1HnNraI=", "wjKMAVW/X4cC8VfboYc2Jxn9l72OJX/gQeNxWRSfRlM=", "Foyb9H1nNdEN/TGIleVYO4D5kxSFbF+ZK/iACXMKDUA=", "mdkuXA30wGxXi/WWqyaGXShcgt3T1yYW+lrsZLBFFuo=", "rO8wDSOjmY8VkspFqYaJS4TV5HxywICMlHM8gTxXkAA=", "1mfy94KpcItqshH9+gwqV6jccupcaMpVsF28New8zDY=", "vS7O4ozHIQZJWBiov+mkpI27GE8zAmVCEkRcP3NDyNE="], "checkpoint": {"envelope": "rekor.sigstore.dev - 1193050959916656506\n419548542\nwlKXR54PophNQ/0UdItOEjggE4Nmv8UkqyyDMY4pgnM=\n\n\u2014 rekor.sigstore.dev wNI9ajBGAiEAjO2Uzz0UVnH3teQh+RtJ9RLGQSARGAigGqpfi+NgLlACIQDwk9tZVH3hztqNTyMxNjfx+bqqOGGYLDPaBWXUePu8zA==\n"}}, "canonicalizedBody": "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"}], "timestampVerificationData": {}}, "dsseEnvelope": {"payload": "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", "payloadType": "application/vnd.in-toto+json", "signatures": [{"sig": "MEUCIQDunb7/Dbn3lnyweCEY9DC578KQhPy3QzVtcT+fzq30aQIgZJDMfGsiRhohqxaHWHDHmiZjC/JL+tD3qV6m3V4WocE="}]}}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<|end_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|end_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|pad|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|unk|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}