Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
+tokenizer_config.json filter=lfs diff=lfs merge=lfs -text

config.json

@@ -0,0 +1,34 @@
+{
+  "architectures": [
+    "Gemma2ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attn_logit_softcapping": 50.0,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "eos_token_id": 8,
+  "final_logit_softcapping": 30.0,
+  "head_dim": 128,
+  "hidden_act": "gelu_pytorch_tanh",
+  "hidden_activation": "gelu_pytorch_tanh",
+  "hidden_size": 4608,
+  "initializer_range": 0.02,
+  "intermediate_size": 36864,
+  "max_position_embeddings": 8192,
+  "model_type": "gemma2",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 46,
+  "num_key_value_heads": 16,
+  "pad_token_id": 0,
+  "query_pre_attn_scalar": 144,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "sliding_window_size": 4096,
+  "torch_dtype": "float32",
+  "transformers_version": "4.42.0.dev0",
+  "use_cache": true,
+  "vocab_size": 256000,
+  "_attn_implementation": "eager"
+}

generation_config.json

@@ -0,0 +1,8 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "eos_token_id": 8,
+  "pad_token_id": 0,
+  "transformers_version": "4.42.0.dev0"
+}

mergekit_config.yml

@@ -0,0 +1,14 @@
+base_model: IntervitensInc/gemma-2-27b-chatml
+dtype: float32
+merge_method: dare_ties
+models:
+  - model: IntervitensInc/gemma-2-27b-chatml
+  - model: anthracite-core/magnum-v3-27b-KTO-e0.25-r1
+    parameters:
+      weight: 0.5
+  - model: anthracite-core/magnum-v3-27b-KTO-e1-r2
+    parameters:
+      weight: 0.25
+  - model: anthracite-core/magnum-v3-27b-kto-r3
+    parameters:
+      weight: 0.25

model-00001-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3650bc7c6a509baed3bd7c47ec88a8d630cb9a04830ff344f7a0d9ad857c54fb
+size 4718610680

model-00002-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:419fb6fd872cd8abc2dcbb274596a752a5e6d263eb3e52902d8151190f846e36
+size 4529998360

model-00003-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:644a9eceaf0898d10df5fdbe733ef9d846ae2bd41a6502be5ccafaeb39e04251
+size 4529998376

model-00004-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a673a87f17b844366dde03bc0eb2f957d06391e67c984a211ff7602f6c805066
+size 4529998376

model-00005-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:128a7faa2c8cba8ffeccc4f26d0212b595df3499d8caa760163817b00723fba2
+size 4529998376

model-00006-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09ca79709ad7826f3c20460a6efa0429728125fd365b0db4b3b62c6d7059214b
+size 4529998376

model-00007-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d53bf06cbb01d398bed597dd3699d75a6e8bcaba33cba613c97eeae8dfff519
+size 4529998376

model-00008-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84412ac41040272aa18ab7f51d0ef5722966ae5f231c9e44d5aaa6d482352066
+size 4529998368

model-00009-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab3810d8f3dc889b626b1a9e20fda6c70de24c0c60ff457883eacc3f6402833f
+size 4529998376

model-00010-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46648feef79eeecaeed748a826bb00c58e90b2941b1aa0cc46a373dec080f7a8
+size 4529998376

model-00011-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fac510bbddf6c9290c707138887cd8dafbb74da2d0ffa2c8f67d4b403c259fa
+size 4529998376

model-00012-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64bdcb3ffafa187bca13a886e978358002dde7d8918313fabf2da23b2efd0bf9
+size 4529998376

model-00013-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed02d2fac8ecf27b8f8b10e8cab057144bdc80aaf5c8d623ecee9e38c7b6577b
+size 4529998368

model-00014-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcbcf0ca776d0ab087e5207d398112abc2ce585dd36761d692979150a33e113d
+size 4529998376

model-00015-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1527f036ffe3e853c131dd7cae4a2428b8a22a69d79daad35237578e7a5b83f7
+size 4529998376

model-00016-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c21534e6080e3e36ddb4100a4746d1f2cd8bbd8120df5c3469d596ae4a011e15
+size 4529998376

model-00017-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc0799769140fb1b1f0b2176b6e54bef52387a650b107b41f913ba9a0da3438c
+size 4529998376

model-00018-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:812550d6cc4e99d544b986804487e84972d2c45e50860c08b5492fe58f3fe646
+size 4529998376

model-00019-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ed9317afeb511dfb835ee618468931f30b9c2cbfc9a7520dc4442477e0af69c
+size 4529998368

model-00020-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee86fa00e06099aab77510af92ece6e08edcbb01d1bdd19540b6e7fd6b7cae63
+size 4529998376

model-00021-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fd901c6a4c867c330b0222252952e8246461e5038a15087d7649946dfa10ba5
+size 4529998376

model-00022-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79f88b4da2eaf9e8c24b8960e769c45ca96e363a7c0fc4e2f7998b71ada49438
+size 4529998368

model-00023-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbad2222b27c54582d4331edb318fb3ef57110d0aeefd66c2e269c471d8e0403
+size 4529998360

model-00024-of-00024.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:826d15416e1deecb745660da90dcc4b10b7b880791018f6b692a3df437d4f76d
+size 4529998336

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.4.4", "total_size": 108908513280}, "weight_map": {"model.embed_tokens.weight": "model-00001-of-00024.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00024.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00002-of-00024.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00002-of-00024.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00002-of-00024.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00002-of-00024.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00002-of-00024.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00002-of-00024.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00002-of-00024.safetensors", "model.layers.1.input_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00002-of-00024.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00024.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00002-of-00024.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00024.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00002-of-00024.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00002-of-00024.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00002-of-00024.safetensors", "model.layers.10.input_layernorm.weight": "model-00002-of-00024.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00003-of-00024.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00024.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00003-of-00024.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.10.post_feedforward_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.10.pre_feedforward_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00024.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00024.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00024.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00024.safetensors", "model.layers.11.input_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00003-of-00024.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00024.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00003-of-00024.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.11.post_feedforward_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.11.pre_feedforward_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00024.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00024.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00024.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00024.safetensors", "model.layers.12.input_layernorm.weight": "model-00003-of-00024.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00004-of-00024.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00004-of-00024.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00004-of-00024.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.12.post_feedforward_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.12.pre_feedforward_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00004-of-00024.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00004-of-00024.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00004-of-00024.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00004-of-00024.safetensors", "model.layers.13.input_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00004-of-00024.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00004-of-00024.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00004-of-00024.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.13.post_feedforward_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.13.pre_feedforward_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00004-of-00024.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00004-of-00024.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00004-of-00024.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00004-of-00024.safetensors", "model.layers.14.input_layernorm.weight": "model-00004-of-00024.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00005-of-00024.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00005-of-00024.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00005-of-00024.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.14.post_feedforward_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.14.pre_feedforward_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00005-of-00024.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00005-of-00024.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00005-of-00024.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00005-of-00024.safetensors", "model.layers.15.input_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00005-of-00024.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00005-of-00024.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00005-of-00024.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.15.post_feedforward_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.15.pre_feedforward_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00005-of-00024.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00005-of-00024.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00005-of-00024.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00005-of-00024.safetensors", "model.layers.16.input_layernorm.weight": "model-00005-of-00024.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00006-of-00024.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00006-of-00024.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00006-of-00024.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.16.post_feedforward_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.16.pre_feedforward_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00006-of-00024.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00006-of-00024.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00006-of-00024.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00006-of-00024.safetensors", "model.layers.17.input_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00006-of-00024.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00006-of-00024.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00006-of-00024.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.17.post_feedforward_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.17.pre_feedforward_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00006-of-00024.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00006-of-00024.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00006-of-00024.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00006-of-00024.safetensors", "model.layers.18.input_layernorm.weight": "model-00006-of-00024.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00007-of-00024.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00007-of-00024.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00007-of-00024.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.18.post_feedforward_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.18.pre_feedforward_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00007-of-00024.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00007-of-00024.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00007-of-00024.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00007-of-00024.safetensors", "model.layers.19.input_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00007-of-00024.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00007-of-00024.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00007-of-00024.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.19.post_feedforward_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.19.pre_feedforward_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00007-of-00024.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00007-of-00024.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00007-of-00024.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00007-of-00024.safetensors", "model.layers.2.input_layernorm.weight": "model-00007-of-00024.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00008-of-00024.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00008-of-00024.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00008-of-00024.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.2.post_feedforward_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.2.pre_feedforward_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00008-of-00024.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00008-of-00024.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00008-of-00024.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00008-of-00024.safetensors", "model.layers.20.input_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00008-of-00024.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00008-of-00024.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00008-of-00024.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.20.post_feedforward_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.20.pre_feedforward_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00008-of-00024.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00008-of-00024.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00008-of-00024.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00008-of-00024.safetensors", "model.layers.21.input_layernorm.weight": "model-00008-of-00024.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00009-of-00024.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00009-of-00024.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00009-of-00024.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.21.post_feedforward_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.21.pre_feedforward_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00009-of-00024.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00009-of-00024.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00009-of-00024.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00009-of-00024.safetensors", "model.layers.22.input_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00009-of-00024.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00009-of-00024.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00009-of-00024.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.22.post_feedforward_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.22.pre_feedforward_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00009-of-00024.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00009-of-00024.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00009-of-00024.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00009-of-00024.safetensors", "model.layers.23.input_layernorm.weight": "model-00009-of-00024.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00010-of-00024.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00010-of-00024.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00010-of-00024.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.23.post_feedforward_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.23.pre_feedforward_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00010-of-00024.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00010-of-00024.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00010-of-00024.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00010-of-00024.safetensors", "model.layers.24.input_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00010-of-00024.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00010-of-00024.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00010-of-00024.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.24.post_feedforward_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.24.pre_feedforward_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00010-of-00024.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00010-of-00024.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00010-of-00024.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00010-of-00024.safetensors", "model.layers.25.input_layernorm.weight": "model-00010-of-00024.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00011-of-00024.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00011-of-00024.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00011-of-00024.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.25.post_feedforward_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.25.pre_feedforward_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00011-of-00024.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00011-of-00024.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00011-of-00024.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00011-of-00024.safetensors", "model.layers.26.input_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00011-of-00024.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00011-of-00024.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00011-of-00024.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.26.post_feedforward_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.26.pre_feedforward_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00011-of-00024.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00011-of-00024.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00011-of-00024.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00011-of-00024.safetensors", "model.layers.27.input_layernorm.weight": "model-00011-of-00024.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00012-of-00024.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00012-of-00024.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00012-of-00024.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.27.post_feedforward_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.27.pre_feedforward_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00012-of-00024.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00012-of-00024.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00012-of-00024.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00012-of-00024.safetensors", "model.layers.28.input_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00012-of-00024.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00012-of-00024.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00012-of-00024.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.28.post_feedforward_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.28.pre_feedforward_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00012-of-00024.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00012-of-00024.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00012-of-00024.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00012-of-00024.safetensors", "model.layers.29.input_layernorm.weight": "model-00012-of-00024.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00013-of-00024.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00013-of-00024.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00013-of-00024.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.29.post_feedforward_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.29.pre_feedforward_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00013-of-00024.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00013-of-00024.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00013-of-00024.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00013-of-00024.safetensors", "model.layers.3.input_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00013-of-00024.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00013-of-00024.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00013-of-00024.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.3.post_feedforward_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.3.pre_feedforward_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00013-of-00024.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00013-of-00024.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00013-of-00024.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00013-of-00024.safetensors", "model.layers.30.input_layernorm.weight": "model-00013-of-00024.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00014-of-00024.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00014-of-00024.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00014-of-00024.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.30.post_feedforward_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.30.pre_feedforward_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00014-of-00024.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00014-of-00024.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00014-of-00024.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00014-of-00024.safetensors", "model.layers.31.input_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00014-of-00024.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00014-of-00024.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00014-of-00024.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.31.post_feedforward_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.31.pre_feedforward_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00014-of-00024.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00014-of-00024.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00014-of-00024.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00014-of-00024.safetensors", "model.layers.32.input_layernorm.weight": "model-00014-of-00024.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00015-of-00024.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00015-of-00024.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00015-of-00024.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.32.post_feedforward_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.32.pre_feedforward_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00015-of-00024.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00015-of-00024.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00015-of-00024.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00015-of-00024.safetensors", "model.layers.33.input_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00015-of-00024.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00015-of-00024.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00015-of-00024.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.33.post_feedforward_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.33.pre_feedforward_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00015-of-00024.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00015-of-00024.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00015-of-00024.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00015-of-00024.safetensors", "model.layers.34.input_layernorm.weight": "model-00015-of-00024.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00016-of-00024.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00016-of-00024.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00016-of-00024.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.34.post_feedforward_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.34.pre_feedforward_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00016-of-00024.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00016-of-00024.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00016-of-00024.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00016-of-00024.safetensors", "model.layers.35.input_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00016-of-00024.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00016-of-00024.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00016-of-00024.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.35.post_feedforward_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.35.pre_feedforward_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00016-of-00024.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00016-of-00024.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00016-of-00024.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00016-of-00024.safetensors", "model.layers.36.input_layernorm.weight": "model-00016-of-00024.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00017-of-00024.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00017-of-00024.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00017-of-00024.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.36.post_feedforward_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.36.pre_feedforward_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00017-of-00024.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00017-of-00024.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00017-of-00024.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00017-of-00024.safetensors", "model.layers.37.input_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00017-of-00024.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00017-of-00024.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00017-of-00024.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.37.post_feedforward_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.37.pre_feedforward_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00017-of-00024.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00017-of-00024.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00017-of-00024.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00017-of-00024.safetensors", "model.layers.38.input_layernorm.weight": "model-00017-of-00024.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00018-of-00024.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00018-of-00024.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00018-of-00024.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.38.post_feedforward_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.38.pre_feedforward_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00018-of-00024.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00018-of-00024.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00018-of-00024.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00018-of-00024.safetensors", "model.layers.39.input_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00018-of-00024.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00018-of-00024.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00018-of-00024.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.39.post_feedforward_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.39.pre_feedforward_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00018-of-00024.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00018-of-00024.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00018-of-00024.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00018-of-00024.safetensors", "model.layers.4.input_layernorm.weight": "model-00018-of-00024.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00019-of-00024.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00019-of-00024.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00019-of-00024.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.4.post_feedforward_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.4.pre_feedforward_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00019-of-00024.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00019-of-00024.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00019-of-00024.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00019-of-00024.safetensors", "model.layers.40.input_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00019-of-00024.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00019-of-00024.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00019-of-00024.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.40.post_feedforward_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.40.pre_feedforward_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00019-of-00024.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00019-of-00024.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00019-of-00024.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00019-of-00024.safetensors", "model.layers.41.input_layernorm.weight": "model-00019-of-00024.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00020-of-00024.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00020-of-00024.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00020-of-00024.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.41.post_feedforward_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.41.pre_feedforward_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00020-of-00024.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00020-of-00024.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00020-of-00024.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00020-of-00024.safetensors", "model.layers.42.input_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00020-of-00024.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00020-of-00024.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00020-of-00024.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.42.post_feedforward_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.42.pre_feedforward_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00020-of-00024.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00020-of-00024.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00020-of-00024.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00020-of-00024.safetensors", "model.layers.43.input_layernorm.weight": "model-00020-of-00024.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00021-of-00024.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00021-of-00024.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00021-of-00024.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.43.post_feedforward_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.43.pre_feedforward_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00021-of-00024.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00021-of-00024.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00021-of-00024.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00021-of-00024.safetensors", "model.layers.44.input_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00021-of-00024.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00021-of-00024.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00021-of-00024.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.44.post_feedforward_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.44.pre_feedforward_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00021-of-00024.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00021-of-00024.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00021-of-00024.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00021-of-00024.safetensors", "model.layers.45.input_layernorm.weight": "model-00021-of-00024.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00022-of-00024.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00022-of-00024.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00022-of-00024.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.45.post_feedforward_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.45.pre_feedforward_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00022-of-00024.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00022-of-00024.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00022-of-00024.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00022-of-00024.safetensors", "model.layers.5.input_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00022-of-00024.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00022-of-00024.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00022-of-00024.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.5.post_feedforward_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.5.pre_feedforward_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00022-of-00024.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00022-of-00024.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00022-of-00024.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00022-of-00024.safetensors", "model.layers.6.input_layernorm.weight": "model-00022-of-00024.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00023-of-00024.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00023-of-00024.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00023-of-00024.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.6.post_feedforward_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.6.pre_feedforward_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00023-of-00024.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00023-of-00024.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00023-of-00024.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00023-of-00024.safetensors", "model.layers.7.input_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00023-of-00024.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00023-of-00024.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00023-of-00024.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.7.post_feedforward_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.7.pre_feedforward_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00023-of-00024.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00023-of-00024.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00023-of-00024.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00023-of-00024.safetensors", "model.layers.8.input_layernorm.weight": "model-00023-of-00024.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00024-of-00024.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00024-of-00024.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00024-of-00024.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.8.post_feedforward_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.8.pre_feedforward_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00024-of-00024.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00024-of-00024.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00024-of-00024.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00024-of-00024.safetensors", "model.layers.9.input_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00024-of-00024.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00024-of-00024.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00024-of-00024.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.9.post_feedforward_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.9.pre_feedforward_layernorm.weight": "model-00024-of-00024.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00024-of-00024.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00024-of-00024.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00024-of-00024.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00024-of-00024.safetensors", "model.norm.weight": "model-00024-of-00024.safetensors"}}

special_tokens_map.json

@@ -0,0 +1,34 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|im_end|>",
+    "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
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66c28a230d9e5c840e83aeedfcb553006011db12dbc58aa20cc7e515bc1309cc
+size 17518532

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01f545c6099874ce0c5cf6e4ccecd5f2bce689852b576ddf9d49134081c9cb98
+size 4241007

tokenizer_config.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37261851a5317f74bb796a7c5965f67e1e6bfb2c62e37ff6f61d3b90a361b681
+size 17518528