Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ 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

README.md

@@ -0,0 +1,38 @@
+---
+base_model: []
+library_name: transformers
+tags:
+- mergekit
+- merge
+
+---
+# NeonMaid-12B-v2
+
+This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
+
+## Merge Details
+### Merge Method
+
+This model was merged using the [Arcee Fusion](https://arcee.ai) merge method using C:\Users\yamat\Desktop\text-generation-webui\user_data\models\NeonMaid-12B as a base.
+
+### Models Merged
+
+The following models were included in the merge:
+* C:\Users\yamat\Desktop\text-generation-webui\user_data\models\Orihime-Gutenberg-12B
+
+### Configuration
+
+The following YAML configuration was used to produce this model:
+
+```yaml
+base_model: C:\Users\yamat\Desktop\text-generation-webui\user_data\models\NeonMaid-12B
+models:
+  - model: C:\Users\yamat\Desktop\text-generation-webui\user_data\models\Orihime-Gutenberg-12B
+merge_method: arcee_fusion
+dtype: bfloat16
+out_dtype: bfloat16
+parameters:
+  normalize: true
+tokenizer:
+  source: union
+```

config.json

@@ -0,0 +1,26 @@
+{
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 131072,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": false,
+  "vocab_size": 131075
+}

mergekit_config.yml

@@ -0,0 +1,10 @@
+base_model: C:\Users\yamat\Desktop\text-generation-webui\user_data\models\NeonMaid-12B
+models:
+  - model: C:\Users\yamat\Desktop\text-generation-webui\user_data\models\Orihime-Gutenberg-12B
+merge_method: arcee_fusion
+dtype: bfloat16
+out_dtype: bfloat16
+parameters:
+  normalize: true
+tokenizer:
+  source: union

model-00001-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cbfe563a1924be60a0f40a120173083f1cd1fb776e57d114ac4290d9d57d32e
+size 1342208128

model-00002-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e31ada4149d30566b5e2f89734322426fee5a9af0b9fbd2196ac885bc525f05
+size 1342208144

model-00003-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:035b54e1509a90cff965a5b71233de1999363b386cb1e8a9c7f350bdf658e2e6
+size 996189888

model-00004-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a97a767f7fb703472b1b46b0dc6ad97a0c04f72c14dda5affffff443579f12e4
+size 933265104

model-00005-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55d9eee36205fbeeaccc319bb93d9aee87421ab16274c4da254e5c383f2d29b4
+size 943761344

model-00006-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:630415c153dbedaf25bc55ebabf294ccba059c1d039fa6baa170353a37ec22a2
+size 943761344

model-00007-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c422066033458199ebf15fe533b80903a9e78a3090647e854c66eb572e06416
+size 996179560

model-00008-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a94e9d453077b1e8cd511b5dc671759ec35e5f119aa84cb5149e2cab201007a
+size 933265104

model-00009-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:723cadc43fedc76b7f0354112bed6cdff84aa12ead20ebdfa1942d8ae751ea95
+size 943761344

model-00010-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0df7a4ee7d5feb343f43014e3a978e48bee625bc897ba1c51fb3583e03852049
+size 943761344

model-00011-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d67c10b7f386edc1b81ed30ea57f717bc8b062602fd9a2be1bace875f503d7a
+size 996179560

model-00012-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a89bd96f5401d38b48ced72085e7ce84f7d04d8a5a5c80e3fd28b65485ad2488
+size 933265104

model-00013-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:231f740c80fce4fdd396444445d410fb537c30c8663f8c21e4c145bdc07581bb
+size 943761344

model-00014-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1af7c6a053368f0fe7f671eeddbd0d91f634ed6f56bdfd408ec164a21b79c711
+size 943761344

model-00015-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01f075d25a16cbaa710894d34007cc07e9bade2f4b4b041b0cc01bfec20ad6b1
+size 996179560

model-00016-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c671c08e903c42ba0be5d28dffb8c33508a38e126eb05117fab68dedf56e4352
+size 933265096

model-00017-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8644a5b64db8a787c965ad6149168deb9fbf2f0b20f7d54a43055d67852fad1b
+size 943761344

model-00018-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:236c274ec8c8091dcd36479b0bf33437fc9d133c68a33ee533c5881e180d1bc6
+size 943761344

model-00019-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02b2a44ab4a7029063fb6fb9e088bc2460e32672a1b9fe4ab5b89e6e80f68f5c
+size 996179560

model-00020-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e43a5c694a1192c1914e6b80dda17d605d1973e929a21ba7b6da17668e9cd374
+size 933265104

model-00021-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06130cf8a26ee3a227c8ef62ac3d80f074ad95c8ef1a9f6fc58b5743ae16bffe
+size 943761344

model-00022-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a33020438a340adc06b2056c0eaa300dead3a53a84bb0e040d3217b85540ae0
+size 943761336

model-00023-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:420bd1f53fe478629cd4f4dcafffaa654ffe97f5da469e0a3ddcf6b8312d0041
+size 996179552

model-00024-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09403b2b39887c844ef49ac1757392e7facddc5ac283694d52abe6bb2ca04bf2
+size 933265088

model-00025-of-00025.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:938ccc97384d8e6d4ce45927207c65c1f2f9c835687d527f6f1df8576e680c3f
+size 796960560

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.1.2"}, "weight_map": {"lm_head.weight": "model-00001-of-00025.safetensors", "model.embed_tokens.weight": "model-00002-of-00025.safetensors", "model.layers.0.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00003-of-00025.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00003-of-00025.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00003-of-00025.safetensors", "model.layers.1.input_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00003-of-00025.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00003-of-00025.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00003-of-00025.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00003-of-00025.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00003-of-00025.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00004-of-00025.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00004-of-00025.safetensors", "model.layers.11.input_layernorm.weight": "model-00004-of-00025.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00004-of-00025.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00004-of-00025.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.12.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00005-of-00025.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00005-of-00025.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00005-of-00025.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00005-of-00025.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00005-of-00025.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00005-of-00025.safetensors", "model.layers.13.input_layernorm.weight": "model-00005-of-00025.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00005-of-00025.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.14.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00006-of-00025.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00006-of-00025.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00006-of-00025.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00006-of-00025.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00006-of-00025.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00006-of-00025.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00006-of-00025.safetensors", "model.layers.15.input_layernorm.weight": "model-00006-of-00025.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00007-of-00025.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00007-of-00025.safetensors", "model.layers.16.input_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00007-of-00025.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00007-of-00025.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00007-of-00025.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00007-of-00025.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00007-of-00025.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.17.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00008-of-00025.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00008-of-00025.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00008-of-00025.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00008-of-00025.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00008-of-00025.safetensors", "model.layers.18.input_layernorm.weight": "model-00008-of-00025.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00008-of-00025.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00008-of-00025.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.19.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00009-of-00025.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00009-of-00025.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00009-of-00025.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00009-of-00025.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00009-of-00025.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00009-of-00025.safetensors", "model.layers.2.input_layernorm.weight": "model-00009-of-00025.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00009-of-00025.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.20.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00010-of-00025.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00010-of-00025.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00010-of-00025.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00010-of-00025.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00010-of-00025.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00010-of-00025.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00010-of-00025.safetensors", "model.layers.21.input_layernorm.weight": "model-00010-of-00025.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00011-of-00025.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00011-of-00025.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00011-of-00025.safetensors", "model.layers.22.input_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00011-of-00025.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00011-of-00025.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00011-of-00025.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00011-of-00025.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00011-of-00025.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.23.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00012-of-00025.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00012-of-00025.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00012-of-00025.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00012-of-00025.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00012-of-00025.safetensors", "model.layers.24.input_layernorm.weight": "model-00012-of-00025.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00012-of-00025.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00012-of-00025.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.25.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00013-of-00025.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00013-of-00025.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00013-of-00025.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00013-of-00025.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00013-of-00025.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00013-of-00025.safetensors", "model.layers.26.input_layernorm.weight": "model-00013-of-00025.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00013-of-00025.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.27.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00014-of-00025.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00014-of-00025.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00014-of-00025.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00014-of-00025.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00014-of-00025.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00014-of-00025.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00014-of-00025.safetensors", "model.layers.28.input_layernorm.weight": "model-00014-of-00025.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00015-of-00025.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00015-of-00025.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00015-of-00025.safetensors", "model.layers.29.input_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00015-of-00025.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00015-of-00025.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00015-of-00025.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00015-of-00025.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00015-of-00025.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.3.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00016-of-00025.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00016-of-00025.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00016-of-00025.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00016-of-00025.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00016-of-00025.safetensors", "model.layers.30.input_layernorm.weight": "model-00016-of-00025.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00016-of-00025.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00016-of-00025.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.31.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00017-of-00025.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00017-of-00025.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00017-of-00025.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00017-of-00025.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00017-of-00025.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00017-of-00025.safetensors", "model.layers.32.input_layernorm.weight": "model-00017-of-00025.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00017-of-00025.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.33.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00018-of-00025.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00018-of-00025.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00018-of-00025.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00018-of-00025.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00018-of-00025.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00018-of-00025.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00018-of-00025.safetensors", "model.layers.34.input_layernorm.weight": "model-00018-of-00025.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00019-of-00025.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00019-of-00025.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00019-of-00025.safetensors", "model.layers.35.input_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00019-of-00025.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00019-of-00025.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00019-of-00025.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00019-of-00025.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00019-of-00025.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.36.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00020-of-00025.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00020-of-00025.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00020-of-00025.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00020-of-00025.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00020-of-00025.safetensors", "model.layers.37.input_layernorm.weight": "model-00020-of-00025.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00020-of-00025.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00020-of-00025.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.38.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00021-of-00025.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00021-of-00025.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00021-of-00025.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00021-of-00025.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00021-of-00025.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00021-of-00025.safetensors", "model.layers.39.input_layernorm.weight": "model-00021-of-00025.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00021-of-00025.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.4.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00022-of-00025.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00022-of-00025.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00022-of-00025.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00022-of-00025.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00022-of-00025.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00022-of-00025.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00022-of-00025.safetensors", "model.layers.5.input_layernorm.weight": "model-00022-of-00025.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00023-of-00025.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00023-of-00025.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00023-of-00025.safetensors", "model.layers.6.input_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00023-of-00025.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00023-of-00025.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00023-of-00025.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00023-of-00025.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00023-of-00025.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00024-of-00025.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.7.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00024-of-00025.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00024-of-00025.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00024-of-00025.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00024-of-00025.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00024-of-00025.safetensors", "model.layers.8.input_layernorm.weight": "model-00024-of-00025.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00024-of-00025.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00024-of-00025.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00025-of-00025.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00025-of-00025.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00025-of-00025.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00025-of-00025.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00025-of-00025.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00025-of-00025.safetensors", "model.layers.9.input_layernorm.weight": "model-00025-of-00025.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00025-of-00025.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00025-of-00025.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00025-of-00025.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00025-of-00025.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00025-of-00025.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00025-of-00025.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00025-of-00025.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00025-of-00025.safetensors", "model.norm.weight": "model-00025-of-00025.safetensors"}}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<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 sha256:33cb74e9ca1c0323b1be5f7e367b96bae218c22d0c4ed5b0edeec39f3a8755c3
+size 17078856