Upload folder using huggingface_hub

README.md

@@ -0,0 +1,41 @@
+---
+base_model:
+- sophosympatheia/Midnight-Miqu-70B-v1.5
+- BeaverAI/Moist-Miqu-70B-v1-RC1
+library_name: transformers
+tags:
+- mergekit
+- merge
+
+---
+# merged
+
+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 SLERP merge method.
+
+### Models Merged
+
+The following models were included in the merge:
+* [sophosympatheia/Midnight-Miqu-70B-v1.5](https://huggingface.co/sophosympatheia/Midnight-Miqu-70B-v1.5)
+* [BeaverAI/Moist-Miqu-70B-v1-RC1](https://huggingface.co/BeaverAI/Moist-Miqu-70B-v1-RC1)
+
+### Configuration
+
+The following YAML configuration was used to produce this model:
+
+```yaml
+models:
+  - model: BeaverAI/Moist-Miqu-70B-v1-RC1
+  - model: sophosympatheia/Midnight-Miqu-70B-v1.5
+merge_method: slerp
+base_model: BeaverAI/Moist-Miqu-70B-v1-RC1
+parameters:
+  t:
+    - value: 0.60
+dtype: float16
+
+```

config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "BeaverAI/Moist-Miqu-70B-v1-RC1",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "intermediate_size": 28672,
+  "max_position_embeddings": 32764,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 80,
+  "num_key_value_heads": 8,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 1000000,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.41.2",
+  "use_cache": true,
+  "vocab_size": 32000
+}

mergekit_config.yml

@@ -0,0 +1,9 @@
+models:
+  - model: BeaverAI/Moist-Miqu-70B-v1-RC1
+  - model: sophosympatheia/Midnight-Miqu-70B-v1.5
+merge_method: slerp
+base_model: BeaverAI/Moist-Miqu-70B-v1-RC1
+parameters:
+  t:
+    - value: 0.60
+dtype: float16

model-00001-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4a80bf384a33aa7594ce936f5913f927945dae7f6864d28bd81bebbbb736ecc
+size 4940974552

model-00002-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1cc83f1b15215aed71d0e523aa8427b23b7f4b8902620313e48b648262d9e764
+size 4664167392

model-00003-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a5a10afe1181209b9f65ddfb4cc19ee978696e37f86002ea8220a5dacb85535
+size 4982917880

model-00004-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3bbfbd25f7eacf2600117429666dba87408b70d47071478f1b886064396e16b
+size 4999711704

model-00005-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eaa0d58691b657e4052cabdff5877c1b96a1c47faa418302b8d7a69f1f11b3c8
+size 4647373544

model-00006-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bcb846a7f597c11649a46107ad04c629a7a4605941b02c88aa393ca02995030d
+size 4664167384

model-00007-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3cc794970cc0ea635148d82817eabea057f29629576da930f7505cd32a9ec7c
+size 4664167392

model-00008-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfbd259c6c7ed3ac26ac32926d7d7da9b89e9dd0e826765a01735db17dd73f9d
+size 4982917880

model-00009-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38ff831ca0704d02a801a96aa10df3cfc2d0b0510da7d8dc0db869cbf78a79d4
+size 4999711696

model-00010-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9c7cd931a85292e724f46072affdd9fb641c9e6e484b4e1a742da19590375a5
+size 4647373552

model-00011-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb8f325226b6bac3d78ee9f044e19f0ea0cb83fccf6cc12da0ad3580901a1c5a
+size 4664167384

model-00012-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dc89beb18a9cbffb2cd4b1e4f66546f0a0af5b23ad8fe783a5c6c5058f288e3
+size 4664167392

model-00013-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:076d99de602ed39bb0990ee1cc6b851956c8816df058895c57a919e449913b9f
+size 4982917872

model-00014-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee9a0452a37bcc542966bab2dbc02e1f299b1dc0b7991fb990202c4a578b73c8
+size 4999711704

model-00015-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64424dd30a1e832eef1013c5a992489f317ba4ac76fe447967454cb26ed6d567
+size 4647373552

model-00016-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d82bc1f76b1b84c017d2f80f865a0fcb7e30e801f087df248ec968953d299830
+size 4664167384

model-00017-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17ff10c999c791b8fe4580ec54a3fdf75f6cb874c7af3b53e7f78e51f6dbbf67
+size 4664167384

model-00018-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6def55171fb1a7e90b7deb35804c628b21d7727b1250af48091023835dc53e9
+size 4982917880

model-00019-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c67e648f0da0ea0402bdb39d079abe9161fd73401535cff0556857425001428
+size 4999711704

model-00020-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bad737c774ea8b270285f29b39d3e84ce1f05cce1e54e89668809a11aaf26f91
+size 4647373552

model-00021-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04a3ba596dd3fbc44c2267f73995046f3b43c91c5227ec63cb6d03ed7b3d624f
+size 4664167376

model-00022-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8340606c419bd032ec38dd5233aa4edd2f721442fd28a276fa68fc1178cd380
+size 4664167392

model-00023-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70863f6f8abfb4a374f707caa6da22b7eb7e12ffa78f997e175cefd0b8f3a0a0
+size 4982917880

model-00024-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dade4d0ac306d07153b8d66f1d7123fa7e9a1bcebecdbc190f34c752b5a70aa3
+size 4999711704

model-00025-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1d18794f6e66824baf726428bdf91aaed6d717ab44a9631c6cd4f1b7edfd9f5
+size 4647373544

model-00026-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d14a242e47132d1742e34edc2538f6a77f48ff8524be8e2cc31158084f74daba
+size 4664167384

model-00027-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:832a94433e925e5bc9a83971d598f2b1189cd2bf062909fd195baf7360519a50
+size 4664167392

model-00028-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0d2ee689066930d7d375cb2b78ef50214ce54e2b63e13b0542ca25c1be611fe
+size 4982917880

model-00029-of-00029.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b3905b97bd6d37cc805da62d696603c238d42783708395c11bbb3dd15bc5464
+size 3573631344

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.4.2", "total_size": 137953296384}, "weight_map": {"lm_head.weight": "model-00001-of-00029.safetensors", "model.embed_tokens.weight": "model-00001-of-00029.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00029.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00001-of-00029.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00029.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00001-of-00029.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00029.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00029.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00029.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00029.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00029.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00029.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00029.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00029.safetensors", "model.layers.10.input_layernorm.weight": "model-00001-of-00029.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00001-of-00029.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00029.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00002-of-00029.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00029.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00029.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00029.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00029.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.input_layernorm.weight": "model-00002-of-00029.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00029.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00029.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.input_layernorm.weight": "model-00002-of-00029.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00029.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00029.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00029.safetensors", "model.layers.13.input_layernorm.weight": "model-00002-of-00029.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00003-of-00029.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00029.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00003-of-00029.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00003-of-00029.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00029.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00029.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00029.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.input_layernorm.weight": "model-00003-of-00029.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00003-of-00029.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00003-of-00029.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00003-of-00029.safetensors", "model.layers.15.input_layernorm.weight": "model-00003-of-00029.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00003-of-00029.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00003-of-00029.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00003-of-00029.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00003-of-00029.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00003-of-00029.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00003-of-00029.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00029.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.input_layernorm.weight": "model-00004-of-00029.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00029.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00029.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.input_layernorm.weight": "model-00004-of-00029.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00004-of-00029.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00029.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00029.safetensors", "model.layers.18.input_layernorm.weight": "model-00004-of-00029.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00004-of-00029.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00004-of-00029.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00004-of-00029.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00004-of-00029.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00004-of-00029.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00005-of-00029.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00029.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.input_layernorm.weight": "model-00005-of-00029.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00029.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00029.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.input_layernorm.weight": "model-00005-of-00029.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00005-of-00029.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00005-of-00029.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00005-of-00029.safetensors", "model.layers.20.input_layernorm.weight": "model-00005-of-00029.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00005-of-00029.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00029.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00006-of-00029.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00006-of-00029.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00006-of-00029.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00006-of-00029.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00006-of-00029.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.input_layernorm.weight": "model-00006-of-00029.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00006-of-00029.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00006-of-00029.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.input_layernorm.weight": "model-00006-of-00029.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00006-of-00029.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00006-of-00029.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00006-of-00029.safetensors", "model.layers.23.input_layernorm.weight": "model-00006-of-00029.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00006-of-00029.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00007-of-00029.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00007-of-00029.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00007-of-00029.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00007-of-00029.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00007-of-00029.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00007-of-00029.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.input_layernorm.weight": "model-00007-of-00029.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00007-of-00029.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00007-of-00029.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.input_layernorm.weight": "model-00007-of-00029.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00007-of-00029.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00007-of-00029.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00007-of-00029.safetensors", "model.layers.26.input_layernorm.weight": "model-00007-of-00029.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00008-of-00029.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00008-of-00029.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00008-of-00029.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00008-of-00029.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00008-of-00029.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00008-of-00029.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00008-of-00029.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.input_layernorm.weight": "model-00008-of-00029.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00008-of-00029.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00008-of-00029.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00008-of-00029.safetensors", "model.layers.28.input_layernorm.weight": "model-00008-of-00029.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00008-of-00029.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00008-of-00029.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00008-of-00029.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00008-of-00029.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00008-of-00029.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00008-of-00029.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00009-of-00029.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.input_layernorm.weight": "model-00009-of-00029.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00009-of-00029.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00009-of-00029.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.input_layernorm.weight": "model-00009-of-00029.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00009-of-00029.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00009-of-00029.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00009-of-00029.safetensors", "model.layers.30.input_layernorm.weight": "model-00009-of-00029.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00009-of-00029.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00009-of-00029.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00009-of-00029.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00009-of-00029.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00009-of-00029.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00010-of-00029.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00010-of-00029.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.input_layernorm.weight": "model-00010-of-00029.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00010-of-00029.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00010-of-00029.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.input_layernorm.weight": "model-00010-of-00029.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00010-of-00029.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00010-of-00029.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00010-of-00029.safetensors", "model.layers.33.input_layernorm.weight": "model-00010-of-00029.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00010-of-00029.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00010-of-00029.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00011-of-00029.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00011-of-00029.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00011-of-00029.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00011-of-00029.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00011-of-00029.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.input_layernorm.weight": "model-00011-of-00029.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00011-of-00029.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00011-of-00029.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.input_layernorm.weight": "model-00011-of-00029.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00011-of-00029.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00011-of-00029.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00011-of-00029.safetensors", "model.layers.36.input_layernorm.weight": "model-00011-of-00029.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00011-of-00029.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00012-of-00029.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00012-of-00029.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00012-of-00029.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00012-of-00029.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00012-of-00029.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00012-of-00029.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.input_layernorm.weight": "model-00012-of-00029.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00012-of-00029.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00012-of-00029.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.input_layernorm.weight": "model-00012-of-00029.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00012-of-00029.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00012-of-00029.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00012-of-00029.safetensors", "model.layers.39.input_layernorm.weight": "model-00012-of-00029.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00013-of-00029.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00013-of-00029.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00013-of-00029.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00013-of-00029.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00013-of-00029.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00013-of-00029.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00013-of-00029.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.input_layernorm.weight": "model-00013-of-00029.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00013-of-00029.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00013-of-00029.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00013-of-00029.safetensors", "model.layers.40.input_layernorm.weight": "model-00013-of-00029.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00013-of-00029.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00013-of-00029.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00013-of-00029.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00013-of-00029.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00013-of-00029.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00013-of-00029.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00014-of-00029.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.input_layernorm.weight": "model-00014-of-00029.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00014-of-00029.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00014-of-00029.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.input_layernorm.weight": "model-00014-of-00029.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00014-of-00029.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00014-of-00029.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00014-of-00029.safetensors", "model.layers.43.input_layernorm.weight": "model-00014-of-00029.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00014-of-00029.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00014-of-00029.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00014-of-00029.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00014-of-00029.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00014-of-00029.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00015-of-00029.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00015-of-00029.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.input_layernorm.weight": "model-00015-of-00029.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00015-of-00029.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00015-of-00029.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.input_layernorm.weight": "model-00015-of-00029.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00015-of-00029.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00015-of-00029.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00015-of-00029.safetensors", "model.layers.46.input_layernorm.weight": "model-00015-of-00029.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00015-of-00029.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00015-of-00029.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00016-of-00029.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00016-of-00029.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00016-of-00029.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00016-of-00029.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00016-of-00029.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.input_layernorm.weight": "model-00016-of-00029.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00016-of-00029.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00016-of-00029.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.input_layernorm.weight": "model-00016-of-00029.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00016-of-00029.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00016-of-00029.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00016-of-00029.safetensors", "model.layers.49.input_layernorm.weight": "model-00016-of-00029.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00016-of-00029.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00017-of-00029.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00017-of-00029.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00017-of-00029.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00017-of-00029.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00017-of-00029.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00017-of-00029.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.input_layernorm.weight": "model-00017-of-00029.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00017-of-00029.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00017-of-00029.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.input_layernorm.weight": "model-00017-of-00029.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00017-of-00029.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00017-of-00029.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00017-of-00029.safetensors", "model.layers.51.input_layernorm.weight": "model-00017-of-00029.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00018-of-00029.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00018-of-00029.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00018-of-00029.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00018-of-00029.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00018-of-00029.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00018-of-00029.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00018-of-00029.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.input_layernorm.weight": "model-00018-of-00029.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00018-of-00029.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00018-of-00029.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00018-of-00029.safetensors", "model.layers.53.input_layernorm.weight": "model-00018-of-00029.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00018-of-00029.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00018-of-00029.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00018-of-00029.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00018-of-00029.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00018-of-00029.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00018-of-00029.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00019-of-00029.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.input_layernorm.weight": "model-00019-of-00029.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00019-of-00029.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00019-of-00029.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.input_layernorm.weight": "model-00019-of-00029.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00019-of-00029.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00019-of-00029.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00019-of-00029.safetensors", "model.layers.56.input_layernorm.weight": "model-00019-of-00029.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00019-of-00029.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00019-of-00029.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00019-of-00029.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00019-of-00029.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00019-of-00029.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00020-of-00029.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00020-of-00029.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.input_layernorm.weight": "model-00020-of-00029.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00020-of-00029.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00020-of-00029.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.input_layernorm.weight": "model-00020-of-00029.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00020-of-00029.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00020-of-00029.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00020-of-00029.safetensors", "model.layers.59.input_layernorm.weight": "model-00020-of-00029.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00020-of-00029.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00020-of-00029.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00021-of-00029.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00021-of-00029.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00021-of-00029.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00021-of-00029.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00021-of-00029.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.input_layernorm.weight": "model-00021-of-00029.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00021-of-00029.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00021-of-00029.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.input_layernorm.weight": "model-00021-of-00029.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00021-of-00029.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00021-of-00029.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00021-of-00029.safetensors", "model.layers.61.input_layernorm.weight": "model-00021-of-00029.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00021-of-00029.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00022-of-00029.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00022-of-00029.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00022-of-00029.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00022-of-00029.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00022-of-00029.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00022-of-00029.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.input_layernorm.weight": "model-00022-of-00029.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00022-of-00029.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00022-of-00029.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.input_layernorm.weight": "model-00022-of-00029.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00022-of-00029.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00022-of-00029.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00022-of-00029.safetensors", "model.layers.64.input_layernorm.weight": "model-00022-of-00029.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00023-of-00029.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00023-of-00029.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00023-of-00029.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00023-of-00029.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00023-of-00029.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00023-of-00029.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00023-of-00029.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.input_layernorm.weight": "model-00023-of-00029.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00023-of-00029.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00023-of-00029.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00023-of-00029.safetensors", "model.layers.66.input_layernorm.weight": "model-00023-of-00029.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00023-of-00029.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00023-of-00029.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00023-of-00029.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00023-of-00029.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00023-of-00029.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00023-of-00029.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00024-of-00029.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.input_layernorm.weight": "model-00024-of-00029.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00024-of-00029.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00024-of-00029.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.input_layernorm.weight": "model-00024-of-00029.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00024-of-00029.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00024-of-00029.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00024-of-00029.safetensors", "model.layers.69.input_layernorm.weight": "model-00024-of-00029.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00024-of-00029.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00024-of-00029.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00024-of-00029.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00024-of-00029.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00024-of-00029.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00025-of-00029.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00025-of-00029.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.input_layernorm.weight": "model-00025-of-00029.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00025-of-00029.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00025-of-00029.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.input_layernorm.weight": "model-00025-of-00029.safetensors", "model.layers.70.mlp.down_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.mlp.gate_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.mlp.up_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.post_attention_layernorm.weight": "model-00025-of-00029.safetensors", "model.layers.70.self_attn.k_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.self_attn.o_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.self_attn.q_proj.weight": "model-00025-of-00029.safetensors", "model.layers.70.self_attn.v_proj.weight": "model-00025-of-00029.safetensors", "model.layers.71.input_layernorm.weight": "model-00025-of-00029.safetensors", "model.layers.71.mlp.down_proj.weight": "model-00025-of-00029.safetensors", "model.layers.71.mlp.gate_proj.weight": "model-00025-of-00029.safetensors", "model.layers.71.mlp.up_proj.weight": "model-00026-of-00029.safetensors", "model.layers.71.post_attention_layernorm.weight": "model-00026-of-00029.safetensors", "model.layers.71.self_attn.k_proj.weight": "model-00026-of-00029.safetensors", "model.layers.71.self_attn.o_proj.weight": "model-00026-of-00029.safetensors", "model.layers.71.self_attn.q_proj.weight": "model-00026-of-00029.safetensors", "model.layers.71.self_attn.v_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.input_layernorm.weight": "model-00026-of-00029.safetensors", "model.layers.72.mlp.down_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.mlp.gate_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.mlp.up_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.post_attention_layernorm.weight": "model-00026-of-00029.safetensors", "model.layers.72.self_attn.k_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.self_attn.o_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.self_attn.q_proj.weight": "model-00026-of-00029.safetensors", "model.layers.72.self_attn.v_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.input_layernorm.weight": "model-00026-of-00029.safetensors", "model.layers.73.mlp.down_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.mlp.gate_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.mlp.up_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.post_attention_layernorm.weight": "model-00026-of-00029.safetensors", "model.layers.73.self_attn.k_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.self_attn.o_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.self_attn.q_proj.weight": "model-00026-of-00029.safetensors", "model.layers.73.self_attn.v_proj.weight": "model-00026-of-00029.safetensors", "model.layers.74.input_layernorm.weight": "model-00026-of-00029.safetensors", "model.layers.74.mlp.down_proj.weight": "model-00026-of-00029.safetensors", "model.layers.74.mlp.gate_proj.weight": "model-00027-of-00029.safetensors", "model.layers.74.mlp.up_proj.weight": "model-00027-of-00029.safetensors", "model.layers.74.post_attention_layernorm.weight": "model-00027-of-00029.safetensors", "model.layers.74.self_attn.k_proj.weight": "model-00027-of-00029.safetensors", "model.layers.74.self_attn.o_proj.weight": "model-00027-of-00029.safetensors", "model.layers.74.self_attn.q_proj.weight": "model-00027-of-00029.safetensors", "model.layers.74.self_attn.v_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.input_layernorm.weight": "model-00027-of-00029.safetensors", "model.layers.75.mlp.down_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.mlp.gate_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.mlp.up_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.post_attention_layernorm.weight": "model-00027-of-00029.safetensors", "model.layers.75.self_attn.k_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.self_attn.o_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.self_attn.q_proj.weight": "model-00027-of-00029.safetensors", "model.layers.75.self_attn.v_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.input_layernorm.weight": "model-00027-of-00029.safetensors", "model.layers.76.mlp.down_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.mlp.gate_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.mlp.up_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.post_attention_layernorm.weight": "model-00027-of-00029.safetensors", "model.layers.76.self_attn.k_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.self_attn.o_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.self_attn.q_proj.weight": "model-00027-of-00029.safetensors", "model.layers.76.self_attn.v_proj.weight": "model-00027-of-00029.safetensors", "model.layers.77.input_layernorm.weight": "model-00027-of-00029.safetensors", "model.layers.77.mlp.down_proj.weight": "model-00028-of-00029.safetensors", "model.layers.77.mlp.gate_proj.weight": "model-00028-of-00029.safetensors", "model.layers.77.mlp.up_proj.weight": "model-00028-of-00029.safetensors", "model.layers.77.post_attention_layernorm.weight": "model-00028-of-00029.safetensors", "model.layers.77.self_attn.k_proj.weight": "model-00028-of-00029.safetensors", "model.layers.77.self_attn.o_proj.weight": "model-00028-of-00029.safetensors", "model.layers.77.self_attn.q_proj.weight": "model-00028-of-00029.safetensors", "model.layers.77.self_attn.v_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.input_layernorm.weight": "model-00028-of-00029.safetensors", "model.layers.78.mlp.down_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.mlp.gate_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.mlp.up_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.post_attention_layernorm.weight": "model-00028-of-00029.safetensors", "model.layers.78.self_attn.k_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.self_attn.o_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.self_attn.q_proj.weight": "model-00028-of-00029.safetensors", "model.layers.78.self_attn.v_proj.weight": "model-00028-of-00029.safetensors", "model.layers.79.input_layernorm.weight": "model-00028-of-00029.safetensors", "model.layers.79.mlp.down_proj.weight": "model-00028-of-00029.safetensors", "model.layers.79.mlp.gate_proj.weight": "model-00028-of-00029.safetensors", "model.layers.79.mlp.up_proj.weight": "model-00028-of-00029.safetensors", "model.layers.79.post_attention_layernorm.weight": "model-00028-of-00029.safetensors", "model.layers.79.self_attn.k_proj.weight": "model-00028-of-00029.safetensors", "model.layers.79.self_attn.o_proj.weight": "model-00028-of-00029.safetensors", "model.layers.79.self_attn.q_proj.weight": "model-00029-of-00029.safetensors", "model.layers.79.self_attn.v_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.input_layernorm.weight": "model-00029-of-00029.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00029-of-00029.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00029-of-00029.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.input_layernorm.weight": "model-00029-of-00029.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00029-of-00029.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00029-of-00029.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00029-of-00029.safetensors", "model.norm.weight": "model-00029-of-00029.safetensors"}}

special_tokens_map.json

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

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,42 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<unk>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}