Upload 8 files
Browse files- .gitattributes +1 -0
- README.md +8 -3
- config.json +1033 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
- model.safetensors.index.json +750 -0
- special_tokens_map.json +30 -0
- tokenizer.json +3 -0
- tokenizer_config.json +1037 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
|
@@ -1,3 +1,8 @@
|
|
| 1 |
-
---
|
| 2 |
-
license:
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
base_model: ByteDance-Seed/Seed-Coder-8B-Instruct
|
| 4 |
+
pipeline_tag: text-generation
|
| 5 |
+
library_name: mlx
|
| 6 |
+
tags:
|
| 7 |
+
- mlx
|
| 8 |
+
---
|
config.json
ADDED
|
@@ -0,0 +1,1033 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 4096,
|
| 11 |
+
"initializer_range": 0.009882118,
|
| 12 |
+
"intermediate_size": 14336,
|
| 13 |
+
"layer_norm_eps": null,
|
| 14 |
+
"max_position_embeddings": 32768,
|
| 15 |
+
"mlp_bias": false,
|
| 16 |
+
"model_type": "llama",
|
| 17 |
+
"num_attention_heads": 32,
|
| 18 |
+
"num_hidden_layers": 32,
|
| 19 |
+
"num_key_value_heads": 8,
|
| 20 |
+
"quantization": {
|
| 21 |
+
"group_size": 64,
|
| 22 |
+
"bits": 4,
|
| 23 |
+
"model.embed_tokens": {
|
| 24 |
+
"group_size": 32,
|
| 25 |
+
"bits": 4
|
| 26 |
+
},
|
| 27 |
+
"model.layers.0.self_attn.q_proj": {
|
| 28 |
+
"group_size": 64,
|
| 29 |
+
"bits": 4
|
| 30 |
+
},
|
| 31 |
+
"model.layers.0.self_attn.k_proj": {
|
| 32 |
+
"group_size": 64,
|
| 33 |
+
"bits": 4
|
| 34 |
+
},
|
| 35 |
+
"model.layers.0.self_attn.v_proj": {
|
| 36 |
+
"group_size": 64,
|
| 37 |
+
"bits": 4
|
| 38 |
+
},
|
| 39 |
+
"model.layers.0.self_attn.o_proj": {
|
| 40 |
+
"group_size": 64,
|
| 41 |
+
"bits": 4
|
| 42 |
+
},
|
| 43 |
+
"model.layers.0.self_attn.rope": false,
|
| 44 |
+
"model.layers.0.mlp.gate_proj": {
|
| 45 |
+
"group_size": 64,
|
| 46 |
+
"bits": 4
|
| 47 |
+
},
|
| 48 |
+
"model.layers.0.mlp.down_proj": {
|
| 49 |
+
"group_size": 64,
|
| 50 |
+
"bits": 4
|
| 51 |
+
},
|
| 52 |
+
"model.layers.0.mlp.up_proj": {
|
| 53 |
+
"group_size": 64,
|
| 54 |
+
"bits": 4
|
| 55 |
+
},
|
| 56 |
+
"model.layers.0.input_layernorm": false,
|
| 57 |
+
"model.layers.0.post_attention_layernorm": false,
|
| 58 |
+
"model.layers.1.self_attn.q_proj": {
|
| 59 |
+
"group_size": 64,
|
| 60 |
+
"bits": 4
|
| 61 |
+
},
|
| 62 |
+
"model.layers.1.self_attn.k_proj": {
|
| 63 |
+
"group_size": 64,
|
| 64 |
+
"bits": 4
|
| 65 |
+
},
|
| 66 |
+
"model.layers.1.self_attn.v_proj": {
|
| 67 |
+
"group_size": 64,
|
| 68 |
+
"bits": 4
|
| 69 |
+
},
|
| 70 |
+
"model.layers.1.self_attn.o_proj": {
|
| 71 |
+
"group_size": 64,
|
| 72 |
+
"bits": 4
|
| 73 |
+
},
|
| 74 |
+
"model.layers.1.self_attn.rope": false,
|
| 75 |
+
"model.layers.1.mlp.gate_proj": {
|
| 76 |
+
"group_size": 64,
|
| 77 |
+
"bits": 4
|
| 78 |
+
},
|
| 79 |
+
"model.layers.1.mlp.down_proj": {
|
| 80 |
+
"group_size": 64,
|
| 81 |
+
"bits": 4
|
| 82 |
+
},
|
| 83 |
+
"model.layers.1.mlp.up_proj": {
|
| 84 |
+
"group_size": 64,
|
| 85 |
+
"bits": 4
|
| 86 |
+
},
|
| 87 |
+
"model.layers.1.input_layernorm": false,
|
| 88 |
+
"model.layers.1.post_attention_layernorm": false,
|
| 89 |
+
"model.layers.2.self_attn.q_proj": {
|
| 90 |
+
"group_size": 64,
|
| 91 |
+
"bits": 4
|
| 92 |
+
},
|
| 93 |
+
"model.layers.2.self_attn.k_proj": {
|
| 94 |
+
"group_size": 64,
|
| 95 |
+
"bits": 4
|
| 96 |
+
},
|
| 97 |
+
"model.layers.2.self_attn.v_proj": {
|
| 98 |
+
"group_size": 64,
|
| 99 |
+
"bits": 4
|
| 100 |
+
},
|
| 101 |
+
"model.layers.2.self_attn.o_proj": {
|
| 102 |
+
"group_size": 64,
|
| 103 |
+
"bits": 4
|
| 104 |
+
},
|
| 105 |
+
"model.layers.2.self_attn.rope": false,
|
| 106 |
+
"model.layers.2.mlp.gate_proj": {
|
| 107 |
+
"group_size": 64,
|
| 108 |
+
"bits": 4
|
| 109 |
+
},
|
| 110 |
+
"model.layers.2.mlp.down_proj": {
|
| 111 |
+
"group_size": 64,
|
| 112 |
+
"bits": 4
|
| 113 |
+
},
|
| 114 |
+
"model.layers.2.mlp.up_proj": {
|
| 115 |
+
"group_size": 64,
|
| 116 |
+
"bits": 4
|
| 117 |
+
},
|
| 118 |
+
"model.layers.2.input_layernorm": false,
|
| 119 |
+
"model.layers.2.post_attention_layernorm": false,
|
| 120 |
+
"model.layers.3.self_attn.q_proj": {
|
| 121 |
+
"group_size": 64,
|
| 122 |
+
"bits": 4
|
| 123 |
+
},
|
| 124 |
+
"model.layers.3.self_attn.k_proj": {
|
| 125 |
+
"group_size": 64,
|
| 126 |
+
"bits": 4
|
| 127 |
+
},
|
| 128 |
+
"model.layers.3.self_attn.v_proj": {
|
| 129 |
+
"group_size": 64,
|
| 130 |
+
"bits": 4
|
| 131 |
+
},
|
| 132 |
+
"model.layers.3.self_attn.o_proj": {
|
| 133 |
+
"group_size": 64,
|
| 134 |
+
"bits": 4
|
| 135 |
+
},
|
| 136 |
+
"model.layers.3.self_attn.rope": false,
|
| 137 |
+
"model.layers.3.mlp.gate_proj": {
|
| 138 |
+
"group_size": 64,
|
| 139 |
+
"bits": 4
|
| 140 |
+
},
|
| 141 |
+
"model.layers.3.mlp.down_proj": {
|
| 142 |
+
"group_size": 64,
|
| 143 |
+
"bits": 4
|
| 144 |
+
},
|
| 145 |
+
"model.layers.3.mlp.up_proj": {
|
| 146 |
+
"group_size": 64,
|
| 147 |
+
"bits": 4
|
| 148 |
+
},
|
| 149 |
+
"model.layers.3.input_layernorm": false,
|
| 150 |
+
"model.layers.3.post_attention_layernorm": false,
|
| 151 |
+
"model.layers.4.self_attn.q_proj": {
|
| 152 |
+
"group_size": 64,
|
| 153 |
+
"bits": 4
|
| 154 |
+
},
|
| 155 |
+
"model.layers.4.self_attn.k_proj": {
|
| 156 |
+
"group_size": 64,
|
| 157 |
+
"bits": 4
|
| 158 |
+
},
|
| 159 |
+
"model.layers.4.self_attn.v_proj": {
|
| 160 |
+
"group_size": 64,
|
| 161 |
+
"bits": 4
|
| 162 |
+
},
|
| 163 |
+
"model.layers.4.self_attn.o_proj": {
|
| 164 |
+
"group_size": 64,
|
| 165 |
+
"bits": 4
|
| 166 |
+
},
|
| 167 |
+
"model.layers.4.self_attn.rope": false,
|
| 168 |
+
"model.layers.4.mlp.gate_proj": {
|
| 169 |
+
"group_size": 64,
|
| 170 |
+
"bits": 4
|
| 171 |
+
},
|
| 172 |
+
"model.layers.4.mlp.down_proj": {
|
| 173 |
+
"group_size": 64,
|
| 174 |
+
"bits": 4
|
| 175 |
+
},
|
| 176 |
+
"model.layers.4.mlp.up_proj": {
|
| 177 |
+
"group_size": 64,
|
| 178 |
+
"bits": 4
|
| 179 |
+
},
|
| 180 |
+
"model.layers.4.input_layernorm": false,
|
| 181 |
+
"model.layers.4.post_attention_layernorm": false,
|
| 182 |
+
"model.layers.5.self_attn.q_proj": {
|
| 183 |
+
"group_size": 64,
|
| 184 |
+
"bits": 4
|
| 185 |
+
},
|
| 186 |
+
"model.layers.5.self_attn.k_proj": {
|
| 187 |
+
"group_size": 64,
|
| 188 |
+
"bits": 4
|
| 189 |
+
},
|
| 190 |
+
"model.layers.5.self_attn.v_proj": {
|
| 191 |
+
"group_size": 64,
|
| 192 |
+
"bits": 4
|
| 193 |
+
},
|
| 194 |
+
"model.layers.5.self_attn.o_proj": {
|
| 195 |
+
"group_size": 64,
|
| 196 |
+
"bits": 4
|
| 197 |
+
},
|
| 198 |
+
"model.layers.5.self_attn.rope": false,
|
| 199 |
+
"model.layers.5.mlp.gate_proj": {
|
| 200 |
+
"group_size": 64,
|
| 201 |
+
"bits": 4
|
| 202 |
+
},
|
| 203 |
+
"model.layers.5.mlp.down_proj": {
|
| 204 |
+
"group_size": 64,
|
| 205 |
+
"bits": 4
|
| 206 |
+
},
|
| 207 |
+
"model.layers.5.mlp.up_proj": {
|
| 208 |
+
"group_size": 64,
|
| 209 |
+
"bits": 4
|
| 210 |
+
},
|
| 211 |
+
"model.layers.5.input_layernorm": false,
|
| 212 |
+
"model.layers.5.post_attention_layernorm": false,
|
| 213 |
+
"model.layers.6.self_attn.q_proj": {
|
| 214 |
+
"group_size": 64,
|
| 215 |
+
"bits": 4
|
| 216 |
+
},
|
| 217 |
+
"model.layers.6.self_attn.k_proj": {
|
| 218 |
+
"group_size": 64,
|
| 219 |
+
"bits": 4
|
| 220 |
+
},
|
| 221 |
+
"model.layers.6.self_attn.v_proj": {
|
| 222 |
+
"group_size": 64,
|
| 223 |
+
"bits": 4
|
| 224 |
+
},
|
| 225 |
+
"model.layers.6.self_attn.o_proj": {
|
| 226 |
+
"group_size": 64,
|
| 227 |
+
"bits": 4
|
| 228 |
+
},
|
| 229 |
+
"model.layers.6.self_attn.rope": false,
|
| 230 |
+
"model.layers.6.mlp.gate_proj": {
|
| 231 |
+
"group_size": 64,
|
| 232 |
+
"bits": 4
|
| 233 |
+
},
|
| 234 |
+
"model.layers.6.mlp.down_proj": {
|
| 235 |
+
"group_size": 64,
|
| 236 |
+
"bits": 4
|
| 237 |
+
},
|
| 238 |
+
"model.layers.6.mlp.up_proj": {
|
| 239 |
+
"group_size": 64,
|
| 240 |
+
"bits": 4
|
| 241 |
+
},
|
| 242 |
+
"model.layers.6.input_layernorm": false,
|
| 243 |
+
"model.layers.6.post_attention_layernorm": false,
|
| 244 |
+
"model.layers.7.self_attn.q_proj": {
|
| 245 |
+
"group_size": 64,
|
| 246 |
+
"bits": 4
|
| 247 |
+
},
|
| 248 |
+
"model.layers.7.self_attn.k_proj": {
|
| 249 |
+
"group_size": 64,
|
| 250 |
+
"bits": 4
|
| 251 |
+
},
|
| 252 |
+
"model.layers.7.self_attn.v_proj": {
|
| 253 |
+
"group_size": 64,
|
| 254 |
+
"bits": 4
|
| 255 |
+
},
|
| 256 |
+
"model.layers.7.self_attn.o_proj": {
|
| 257 |
+
"group_size": 64,
|
| 258 |
+
"bits": 4
|
| 259 |
+
},
|
| 260 |
+
"model.layers.7.self_attn.rope": false,
|
| 261 |
+
"model.layers.7.mlp.gate_proj": {
|
| 262 |
+
"group_size": 64,
|
| 263 |
+
"bits": 4
|
| 264 |
+
},
|
| 265 |
+
"model.layers.7.mlp.down_proj": {
|
| 266 |
+
"group_size": 64,
|
| 267 |
+
"bits": 4
|
| 268 |
+
},
|
| 269 |
+
"model.layers.7.mlp.up_proj": {
|
| 270 |
+
"group_size": 64,
|
| 271 |
+
"bits": 4
|
| 272 |
+
},
|
| 273 |
+
"model.layers.7.input_layernorm": false,
|
| 274 |
+
"model.layers.7.post_attention_layernorm": false,
|
| 275 |
+
"model.layers.8.self_attn.q_proj": {
|
| 276 |
+
"group_size": 64,
|
| 277 |
+
"bits": 4
|
| 278 |
+
},
|
| 279 |
+
"model.layers.8.self_attn.k_proj": {
|
| 280 |
+
"group_size": 64,
|
| 281 |
+
"bits": 4
|
| 282 |
+
},
|
| 283 |
+
"model.layers.8.self_attn.v_proj": {
|
| 284 |
+
"group_size": 64,
|
| 285 |
+
"bits": 4
|
| 286 |
+
},
|
| 287 |
+
"model.layers.8.self_attn.o_proj": {
|
| 288 |
+
"group_size": 64,
|
| 289 |
+
"bits": 4
|
| 290 |
+
},
|
| 291 |
+
"model.layers.8.self_attn.rope": false,
|
| 292 |
+
"model.layers.8.mlp.gate_proj": {
|
| 293 |
+
"group_size": 64,
|
| 294 |
+
"bits": 4
|
| 295 |
+
},
|
| 296 |
+
"model.layers.8.mlp.down_proj": {
|
| 297 |
+
"group_size": 64,
|
| 298 |
+
"bits": 4
|
| 299 |
+
},
|
| 300 |
+
"model.layers.8.mlp.up_proj": {
|
| 301 |
+
"group_size": 64,
|
| 302 |
+
"bits": 4
|
| 303 |
+
},
|
| 304 |
+
"model.layers.8.input_layernorm": false,
|
| 305 |
+
"model.layers.8.post_attention_layernorm": false,
|
| 306 |
+
"model.layers.9.self_attn.q_proj": {
|
| 307 |
+
"group_size": 64,
|
| 308 |
+
"bits": 4
|
| 309 |
+
},
|
| 310 |
+
"model.layers.9.self_attn.k_proj": {
|
| 311 |
+
"group_size": 64,
|
| 312 |
+
"bits": 4
|
| 313 |
+
},
|
| 314 |
+
"model.layers.9.self_attn.v_proj": {
|
| 315 |
+
"group_size": 64,
|
| 316 |
+
"bits": 4
|
| 317 |
+
},
|
| 318 |
+
"model.layers.9.self_attn.o_proj": {
|
| 319 |
+
"group_size": 64,
|
| 320 |
+
"bits": 4
|
| 321 |
+
},
|
| 322 |
+
"model.layers.9.self_attn.rope": false,
|
| 323 |
+
"model.layers.9.mlp.gate_proj": {
|
| 324 |
+
"group_size": 64,
|
| 325 |
+
"bits": 4
|
| 326 |
+
},
|
| 327 |
+
"model.layers.9.mlp.down_proj": {
|
| 328 |
+
"group_size": 64,
|
| 329 |
+
"bits": 4
|
| 330 |
+
},
|
| 331 |
+
"model.layers.9.mlp.up_proj": {
|
| 332 |
+
"group_size": 64,
|
| 333 |
+
"bits": 4
|
| 334 |
+
},
|
| 335 |
+
"model.layers.9.input_layernorm": false,
|
| 336 |
+
"model.layers.9.post_attention_layernorm": false,
|
| 337 |
+
"model.layers.10.self_attn.q_proj": {
|
| 338 |
+
"group_size": 64,
|
| 339 |
+
"bits": 4
|
| 340 |
+
},
|
| 341 |
+
"model.layers.10.self_attn.k_proj": {
|
| 342 |
+
"group_size": 64,
|
| 343 |
+
"bits": 4
|
| 344 |
+
},
|
| 345 |
+
"model.layers.10.self_attn.v_proj": {
|
| 346 |
+
"group_size": 64,
|
| 347 |
+
"bits": 4
|
| 348 |
+
},
|
| 349 |
+
"model.layers.10.self_attn.o_proj": {
|
| 350 |
+
"group_size": 64,
|
| 351 |
+
"bits": 4
|
| 352 |
+
},
|
| 353 |
+
"model.layers.10.self_attn.rope": false,
|
| 354 |
+
"model.layers.10.mlp.gate_proj": {
|
| 355 |
+
"group_size": 64,
|
| 356 |
+
"bits": 4
|
| 357 |
+
},
|
| 358 |
+
"model.layers.10.mlp.down_proj": {
|
| 359 |
+
"group_size": 64,
|
| 360 |
+
"bits": 4
|
| 361 |
+
},
|
| 362 |
+
"model.layers.10.mlp.up_proj": {
|
| 363 |
+
"group_size": 64,
|
| 364 |
+
"bits": 4
|
| 365 |
+
},
|
| 366 |
+
"model.layers.10.input_layernorm": false,
|
| 367 |
+
"model.layers.10.post_attention_layernorm": false,
|
| 368 |
+
"model.layers.11.self_attn.q_proj": {
|
| 369 |
+
"group_size": 64,
|
| 370 |
+
"bits": 4
|
| 371 |
+
},
|
| 372 |
+
"model.layers.11.self_attn.k_proj": {
|
| 373 |
+
"group_size": 64,
|
| 374 |
+
"bits": 4
|
| 375 |
+
},
|
| 376 |
+
"model.layers.11.self_attn.v_proj": {
|
| 377 |
+
"group_size": 64,
|
| 378 |
+
"bits": 4
|
| 379 |
+
},
|
| 380 |
+
"model.layers.11.self_attn.o_proj": {
|
| 381 |
+
"group_size": 64,
|
| 382 |
+
"bits": 4
|
| 383 |
+
},
|
| 384 |
+
"model.layers.11.self_attn.rope": false,
|
| 385 |
+
"model.layers.11.mlp.gate_proj": {
|
| 386 |
+
"group_size": 64,
|
| 387 |
+
"bits": 4
|
| 388 |
+
},
|
| 389 |
+
"model.layers.11.mlp.down_proj": {
|
| 390 |
+
"group_size": 64,
|
| 391 |
+
"bits": 4
|
| 392 |
+
},
|
| 393 |
+
"model.layers.11.mlp.up_proj": {
|
| 394 |
+
"group_size": 64,
|
| 395 |
+
"bits": 4
|
| 396 |
+
},
|
| 397 |
+
"model.layers.11.input_layernorm": false,
|
| 398 |
+
"model.layers.11.post_attention_layernorm": false,
|
| 399 |
+
"model.layers.12.self_attn.q_proj": {
|
| 400 |
+
"group_size": 64,
|
| 401 |
+
"bits": 4
|
| 402 |
+
},
|
| 403 |
+
"model.layers.12.self_attn.k_proj": {
|
| 404 |
+
"group_size": 64,
|
| 405 |
+
"bits": 4
|
| 406 |
+
},
|
| 407 |
+
"model.layers.12.self_attn.v_proj": {
|
| 408 |
+
"group_size": 64,
|
| 409 |
+
"bits": 4
|
| 410 |
+
},
|
| 411 |
+
"model.layers.12.self_attn.o_proj": {
|
| 412 |
+
"group_size": 64,
|
| 413 |
+
"bits": 4
|
| 414 |
+
},
|
| 415 |
+
"model.layers.12.self_attn.rope": false,
|
| 416 |
+
"model.layers.12.mlp.gate_proj": {
|
| 417 |
+
"group_size": 64,
|
| 418 |
+
"bits": 4
|
| 419 |
+
},
|
| 420 |
+
"model.layers.12.mlp.down_proj": {
|
| 421 |
+
"group_size": 64,
|
| 422 |
+
"bits": 4
|
| 423 |
+
},
|
| 424 |
+
"model.layers.12.mlp.up_proj": {
|
| 425 |
+
"group_size": 64,
|
| 426 |
+
"bits": 4
|
| 427 |
+
},
|
| 428 |
+
"model.layers.12.input_layernorm": false,
|
| 429 |
+
"model.layers.12.post_attention_layernorm": false,
|
| 430 |
+
"model.layers.13.self_attn.q_proj": {
|
| 431 |
+
"group_size": 64,
|
| 432 |
+
"bits": 4
|
| 433 |
+
},
|
| 434 |
+
"model.layers.13.self_attn.k_proj": {
|
| 435 |
+
"group_size": 64,
|
| 436 |
+
"bits": 4
|
| 437 |
+
},
|
| 438 |
+
"model.layers.13.self_attn.v_proj": {
|
| 439 |
+
"group_size": 64,
|
| 440 |
+
"bits": 4
|
| 441 |
+
},
|
| 442 |
+
"model.layers.13.self_attn.o_proj": {
|
| 443 |
+
"group_size": 64,
|
| 444 |
+
"bits": 4
|
| 445 |
+
},
|
| 446 |
+
"model.layers.13.self_attn.rope": false,
|
| 447 |
+
"model.layers.13.mlp.gate_proj": {
|
| 448 |
+
"group_size": 64,
|
| 449 |
+
"bits": 4
|
| 450 |
+
},
|
| 451 |
+
"model.layers.13.mlp.down_proj": {
|
| 452 |
+
"group_size": 64,
|
| 453 |
+
"bits": 4
|
| 454 |
+
},
|
| 455 |
+
"model.layers.13.mlp.up_proj": {
|
| 456 |
+
"group_size": 64,
|
| 457 |
+
"bits": 4
|
| 458 |
+
},
|
| 459 |
+
"model.layers.13.input_layernorm": false,
|
| 460 |
+
"model.layers.13.post_attention_layernorm": false,
|
| 461 |
+
"model.layers.14.self_attn.q_proj": {
|
| 462 |
+
"group_size": 64,
|
| 463 |
+
"bits": 4
|
| 464 |
+
},
|
| 465 |
+
"model.layers.14.self_attn.k_proj": {
|
| 466 |
+
"group_size": 64,
|
| 467 |
+
"bits": 4
|
| 468 |
+
},
|
| 469 |
+
"model.layers.14.self_attn.v_proj": {
|
| 470 |
+
"group_size": 64,
|
| 471 |
+
"bits": 4
|
| 472 |
+
},
|
| 473 |
+
"model.layers.14.self_attn.o_proj": {
|
| 474 |
+
"group_size": 64,
|
| 475 |
+
"bits": 4
|
| 476 |
+
},
|
| 477 |
+
"model.layers.14.self_attn.rope": false,
|
| 478 |
+
"model.layers.14.mlp.gate_proj": {
|
| 479 |
+
"group_size": 64,
|
| 480 |
+
"bits": 4
|
| 481 |
+
},
|
| 482 |
+
"model.layers.14.mlp.down_proj": {
|
| 483 |
+
"group_size": 64,
|
| 484 |
+
"bits": 4
|
| 485 |
+
},
|
| 486 |
+
"model.layers.14.mlp.up_proj": {
|
| 487 |
+
"group_size": 64,
|
| 488 |
+
"bits": 4
|
| 489 |
+
},
|
| 490 |
+
"model.layers.14.input_layernorm": false,
|
| 491 |
+
"model.layers.14.post_attention_layernorm": false,
|
| 492 |
+
"model.layers.15.self_attn.q_proj": {
|
| 493 |
+
"group_size": 64,
|
| 494 |
+
"bits": 4
|
| 495 |
+
},
|
| 496 |
+
"model.layers.15.self_attn.k_proj": {
|
| 497 |
+
"group_size": 64,
|
| 498 |
+
"bits": 4
|
| 499 |
+
},
|
| 500 |
+
"model.layers.15.self_attn.v_proj": {
|
| 501 |
+
"group_size": 64,
|
| 502 |
+
"bits": 4
|
| 503 |
+
},
|
| 504 |
+
"model.layers.15.self_attn.o_proj": {
|
| 505 |
+
"group_size": 64,
|
| 506 |
+
"bits": 4
|
| 507 |
+
},
|
| 508 |
+
"model.layers.15.self_attn.rope": false,
|
| 509 |
+
"model.layers.15.mlp.gate_proj": {
|
| 510 |
+
"group_size": 64,
|
| 511 |
+
"bits": 4
|
| 512 |
+
},
|
| 513 |
+
"model.layers.15.mlp.down_proj": {
|
| 514 |
+
"group_size": 64,
|
| 515 |
+
"bits": 4
|
| 516 |
+
},
|
| 517 |
+
"model.layers.15.mlp.up_proj": {
|
| 518 |
+
"group_size": 64,
|
| 519 |
+
"bits": 4
|
| 520 |
+
},
|
| 521 |
+
"model.layers.15.input_layernorm": false,
|
| 522 |
+
"model.layers.15.post_attention_layernorm": false,
|
| 523 |
+
"model.layers.16.self_attn.q_proj": {
|
| 524 |
+
"group_size": 64,
|
| 525 |
+
"bits": 4
|
| 526 |
+
},
|
| 527 |
+
"model.layers.16.self_attn.k_proj": {
|
| 528 |
+
"group_size": 64,
|
| 529 |
+
"bits": 4
|
| 530 |
+
},
|
| 531 |
+
"model.layers.16.self_attn.v_proj": {
|
| 532 |
+
"group_size": 64,
|
| 533 |
+
"bits": 4
|
| 534 |
+
},
|
| 535 |
+
"model.layers.16.self_attn.o_proj": {
|
| 536 |
+
"group_size": 64,
|
| 537 |
+
"bits": 4
|
| 538 |
+
},
|
| 539 |
+
"model.layers.16.self_attn.rope": false,
|
| 540 |
+
"model.layers.16.mlp.gate_proj": {
|
| 541 |
+
"group_size": 64,
|
| 542 |
+
"bits": 4
|
| 543 |
+
},
|
| 544 |
+
"model.layers.16.mlp.down_proj": {
|
| 545 |
+
"group_size": 64,
|
| 546 |
+
"bits": 4
|
| 547 |
+
},
|
| 548 |
+
"model.layers.16.mlp.up_proj": {
|
| 549 |
+
"group_size": 64,
|
| 550 |
+
"bits": 4
|
| 551 |
+
},
|
| 552 |
+
"model.layers.16.input_layernorm": false,
|
| 553 |
+
"model.layers.16.post_attention_layernorm": false,
|
| 554 |
+
"model.layers.17.self_attn.q_proj": {
|
| 555 |
+
"group_size": 64,
|
| 556 |
+
"bits": 4
|
| 557 |
+
},
|
| 558 |
+
"model.layers.17.self_attn.k_proj": {
|
| 559 |
+
"group_size": 64,
|
| 560 |
+
"bits": 4
|
| 561 |
+
},
|
| 562 |
+
"model.layers.17.self_attn.v_proj": {
|
| 563 |
+
"group_size": 64,
|
| 564 |
+
"bits": 4
|
| 565 |
+
},
|
| 566 |
+
"model.layers.17.self_attn.o_proj": {
|
| 567 |
+
"group_size": 64,
|
| 568 |
+
"bits": 4
|
| 569 |
+
},
|
| 570 |
+
"model.layers.17.self_attn.rope": false,
|
| 571 |
+
"model.layers.17.mlp.gate_proj": {
|
| 572 |
+
"group_size": 64,
|
| 573 |
+
"bits": 4
|
| 574 |
+
},
|
| 575 |
+
"model.layers.17.mlp.down_proj": {
|
| 576 |
+
"group_size": 64,
|
| 577 |
+
"bits": 4
|
| 578 |
+
},
|
| 579 |
+
"model.layers.17.mlp.up_proj": {
|
| 580 |
+
"group_size": 64,
|
| 581 |
+
"bits": 4
|
| 582 |
+
},
|
| 583 |
+
"model.layers.17.input_layernorm": false,
|
| 584 |
+
"model.layers.17.post_attention_layernorm": false,
|
| 585 |
+
"model.layers.18.self_attn.q_proj": {
|
| 586 |
+
"group_size": 64,
|
| 587 |
+
"bits": 4
|
| 588 |
+
},
|
| 589 |
+
"model.layers.18.self_attn.k_proj": {
|
| 590 |
+
"group_size": 64,
|
| 591 |
+
"bits": 4
|
| 592 |
+
},
|
| 593 |
+
"model.layers.18.self_attn.v_proj": {
|
| 594 |
+
"group_size": 64,
|
| 595 |
+
"bits": 4
|
| 596 |
+
},
|
| 597 |
+
"model.layers.18.self_attn.o_proj": {
|
| 598 |
+
"group_size": 64,
|
| 599 |
+
"bits": 4
|
| 600 |
+
},
|
| 601 |
+
"model.layers.18.self_attn.rope": false,
|
| 602 |
+
"model.layers.18.mlp.gate_proj": {
|
| 603 |
+
"group_size": 64,
|
| 604 |
+
"bits": 4
|
| 605 |
+
},
|
| 606 |
+
"model.layers.18.mlp.down_proj": {
|
| 607 |
+
"group_size": 64,
|
| 608 |
+
"bits": 4
|
| 609 |
+
},
|
| 610 |
+
"model.layers.18.mlp.up_proj": {
|
| 611 |
+
"group_size": 64,
|
| 612 |
+
"bits": 4
|
| 613 |
+
},
|
| 614 |
+
"model.layers.18.input_layernorm": false,
|
| 615 |
+
"model.layers.18.post_attention_layernorm": false,
|
| 616 |
+
"model.layers.19.self_attn.q_proj": {
|
| 617 |
+
"group_size": 64,
|
| 618 |
+
"bits": 4
|
| 619 |
+
},
|
| 620 |
+
"model.layers.19.self_attn.k_proj": {
|
| 621 |
+
"group_size": 64,
|
| 622 |
+
"bits": 4
|
| 623 |
+
},
|
| 624 |
+
"model.layers.19.self_attn.v_proj": {
|
| 625 |
+
"group_size": 64,
|
| 626 |
+
"bits": 4
|
| 627 |
+
},
|
| 628 |
+
"model.layers.19.self_attn.o_proj": {
|
| 629 |
+
"group_size": 64,
|
| 630 |
+
"bits": 4
|
| 631 |
+
},
|
| 632 |
+
"model.layers.19.self_attn.rope": false,
|
| 633 |
+
"model.layers.19.mlp.gate_proj": {
|
| 634 |
+
"group_size": 64,
|
| 635 |
+
"bits": 4
|
| 636 |
+
},
|
| 637 |
+
"model.layers.19.mlp.down_proj": {
|
| 638 |
+
"group_size": 64,
|
| 639 |
+
"bits": 4
|
| 640 |
+
},
|
| 641 |
+
"model.layers.19.mlp.up_proj": {
|
| 642 |
+
"group_size": 64,
|
| 643 |
+
"bits": 4
|
| 644 |
+
},
|
| 645 |
+
"model.layers.19.input_layernorm": false,
|
| 646 |
+
"model.layers.19.post_attention_layernorm": false,
|
| 647 |
+
"model.layers.20.self_attn.q_proj": {
|
| 648 |
+
"group_size": 64,
|
| 649 |
+
"bits": 4
|
| 650 |
+
},
|
| 651 |
+
"model.layers.20.self_attn.k_proj": {
|
| 652 |
+
"group_size": 64,
|
| 653 |
+
"bits": 4
|
| 654 |
+
},
|
| 655 |
+
"model.layers.20.self_attn.v_proj": {
|
| 656 |
+
"group_size": 64,
|
| 657 |
+
"bits": 4
|
| 658 |
+
},
|
| 659 |
+
"model.layers.20.self_attn.o_proj": {
|
| 660 |
+
"group_size": 64,
|
| 661 |
+
"bits": 4
|
| 662 |
+
},
|
| 663 |
+
"model.layers.20.self_attn.rope": false,
|
| 664 |
+
"model.layers.20.mlp.gate_proj": {
|
| 665 |
+
"group_size": 64,
|
| 666 |
+
"bits": 4
|
| 667 |
+
},
|
| 668 |
+
"model.layers.20.mlp.down_proj": {
|
| 669 |
+
"group_size": 64,
|
| 670 |
+
"bits": 4
|
| 671 |
+
},
|
| 672 |
+
"model.layers.20.mlp.up_proj": {
|
| 673 |
+
"group_size": 64,
|
| 674 |
+
"bits": 4
|
| 675 |
+
},
|
| 676 |
+
"model.layers.20.input_layernorm": false,
|
| 677 |
+
"model.layers.20.post_attention_layernorm": false,
|
| 678 |
+
"model.layers.21.self_attn.q_proj": {
|
| 679 |
+
"group_size": 64,
|
| 680 |
+
"bits": 4
|
| 681 |
+
},
|
| 682 |
+
"model.layers.21.self_attn.k_proj": {
|
| 683 |
+
"group_size": 64,
|
| 684 |
+
"bits": 4
|
| 685 |
+
},
|
| 686 |
+
"model.layers.21.self_attn.v_proj": {
|
| 687 |
+
"group_size": 64,
|
| 688 |
+
"bits": 4
|
| 689 |
+
},
|
| 690 |
+
"model.layers.21.self_attn.o_proj": {
|
| 691 |
+
"group_size": 64,
|
| 692 |
+
"bits": 4
|
| 693 |
+
},
|
| 694 |
+
"model.layers.21.self_attn.rope": false,
|
| 695 |
+
"model.layers.21.mlp.gate_proj": {
|
| 696 |
+
"group_size": 64,
|
| 697 |
+
"bits": 4
|
| 698 |
+
},
|
| 699 |
+
"model.layers.21.mlp.down_proj": {
|
| 700 |
+
"group_size": 64,
|
| 701 |
+
"bits": 4
|
| 702 |
+
},
|
| 703 |
+
"model.layers.21.mlp.up_proj": {
|
| 704 |
+
"group_size": 64,
|
| 705 |
+
"bits": 4
|
| 706 |
+
},
|
| 707 |
+
"model.layers.21.input_layernorm": false,
|
| 708 |
+
"model.layers.21.post_attention_layernorm": false,
|
| 709 |
+
"model.layers.22.self_attn.q_proj": {
|
| 710 |
+
"group_size": 64,
|
| 711 |
+
"bits": 4
|
| 712 |
+
},
|
| 713 |
+
"model.layers.22.self_attn.k_proj": {
|
| 714 |
+
"group_size": 64,
|
| 715 |
+
"bits": 4
|
| 716 |
+
},
|
| 717 |
+
"model.layers.22.self_attn.v_proj": {
|
| 718 |
+
"group_size": 64,
|
| 719 |
+
"bits": 4
|
| 720 |
+
},
|
| 721 |
+
"model.layers.22.self_attn.o_proj": {
|
| 722 |
+
"group_size": 64,
|
| 723 |
+
"bits": 4
|
| 724 |
+
},
|
| 725 |
+
"model.layers.22.self_attn.rope": false,
|
| 726 |
+
"model.layers.22.mlp.gate_proj": {
|
| 727 |
+
"group_size": 64,
|
| 728 |
+
"bits": 4
|
| 729 |
+
},
|
| 730 |
+
"model.layers.22.mlp.down_proj": {
|
| 731 |
+
"group_size": 64,
|
| 732 |
+
"bits": 4
|
| 733 |
+
},
|
| 734 |
+
"model.layers.22.mlp.up_proj": {
|
| 735 |
+
"group_size": 64,
|
| 736 |
+
"bits": 4
|
| 737 |
+
},
|
| 738 |
+
"model.layers.22.input_layernorm": false,
|
| 739 |
+
"model.layers.22.post_attention_layernorm": false,
|
| 740 |
+
"model.layers.23.self_attn.q_proj": {
|
| 741 |
+
"group_size": 64,
|
| 742 |
+
"bits": 4
|
| 743 |
+
},
|
| 744 |
+
"model.layers.23.self_attn.k_proj": {
|
| 745 |
+
"group_size": 64,
|
| 746 |
+
"bits": 4
|
| 747 |
+
},
|
| 748 |
+
"model.layers.23.self_attn.v_proj": {
|
| 749 |
+
"group_size": 64,
|
| 750 |
+
"bits": 4
|
| 751 |
+
},
|
| 752 |
+
"model.layers.23.self_attn.o_proj": {
|
| 753 |
+
"group_size": 64,
|
| 754 |
+
"bits": 4
|
| 755 |
+
},
|
| 756 |
+
"model.layers.23.self_attn.rope": false,
|
| 757 |
+
"model.layers.23.mlp.gate_proj": {
|
| 758 |
+
"group_size": 64,
|
| 759 |
+
"bits": 4
|
| 760 |
+
},
|
| 761 |
+
"model.layers.23.mlp.down_proj": {
|
| 762 |
+
"group_size": 64,
|
| 763 |
+
"bits": 4
|
| 764 |
+
},
|
| 765 |
+
"model.layers.23.mlp.up_proj": {
|
| 766 |
+
"group_size": 64,
|
| 767 |
+
"bits": 4
|
| 768 |
+
},
|
| 769 |
+
"model.layers.23.input_layernorm": false,
|
| 770 |
+
"model.layers.23.post_attention_layernorm": false,
|
| 771 |
+
"model.layers.24.self_attn.q_proj": {
|
| 772 |
+
"group_size": 64,
|
| 773 |
+
"bits": 4
|
| 774 |
+
},
|
| 775 |
+
"model.layers.24.self_attn.k_proj": {
|
| 776 |
+
"group_size": 64,
|
| 777 |
+
"bits": 4
|
| 778 |
+
},
|
| 779 |
+
"model.layers.24.self_attn.v_proj": {
|
| 780 |
+
"group_size": 64,
|
| 781 |
+
"bits": 4
|
| 782 |
+
},
|
| 783 |
+
"model.layers.24.self_attn.o_proj": {
|
| 784 |
+
"group_size": 64,
|
| 785 |
+
"bits": 4
|
| 786 |
+
},
|
| 787 |
+
"model.layers.24.self_attn.rope": false,
|
| 788 |
+
"model.layers.24.mlp.gate_proj": {
|
| 789 |
+
"group_size": 64,
|
| 790 |
+
"bits": 4
|
| 791 |
+
},
|
| 792 |
+
"model.layers.24.mlp.down_proj": {
|
| 793 |
+
"group_size": 64,
|
| 794 |
+
"bits": 4
|
| 795 |
+
},
|
| 796 |
+
"model.layers.24.mlp.up_proj": {
|
| 797 |
+
"group_size": 64,
|
| 798 |
+
"bits": 4
|
| 799 |
+
},
|
| 800 |
+
"model.layers.24.input_layernorm": false,
|
| 801 |
+
"model.layers.24.post_attention_layernorm": false,
|
| 802 |
+
"model.layers.25.self_attn.q_proj": {
|
| 803 |
+
"group_size": 64,
|
| 804 |
+
"bits": 4
|
| 805 |
+
},
|
| 806 |
+
"model.layers.25.self_attn.k_proj": {
|
| 807 |
+
"group_size": 64,
|
| 808 |
+
"bits": 4
|
| 809 |
+
},
|
| 810 |
+
"model.layers.25.self_attn.v_proj": {
|
| 811 |
+
"group_size": 64,
|
| 812 |
+
"bits": 4
|
| 813 |
+
},
|
| 814 |
+
"model.layers.25.self_attn.o_proj": {
|
| 815 |
+
"group_size": 64,
|
| 816 |
+
"bits": 4
|
| 817 |
+
},
|
| 818 |
+
"model.layers.25.self_attn.rope": false,
|
| 819 |
+
"model.layers.25.mlp.gate_proj": {
|
| 820 |
+
"group_size": 64,
|
| 821 |
+
"bits": 4
|
| 822 |
+
},
|
| 823 |
+
"model.layers.25.mlp.down_proj": {
|
| 824 |
+
"group_size": 64,
|
| 825 |
+
"bits": 4
|
| 826 |
+
},
|
| 827 |
+
"model.layers.25.mlp.up_proj": {
|
| 828 |
+
"group_size": 64,
|
| 829 |
+
"bits": 4
|
| 830 |
+
},
|
| 831 |
+
"model.layers.25.input_layernorm": false,
|
| 832 |
+
"model.layers.25.post_attention_layernorm": false,
|
| 833 |
+
"model.layers.26.self_attn.q_proj": {
|
| 834 |
+
"group_size": 64,
|
| 835 |
+
"bits": 4
|
| 836 |
+
},
|
| 837 |
+
"model.layers.26.self_attn.k_proj": {
|
| 838 |
+
"group_size": 64,
|
| 839 |
+
"bits": 4
|
| 840 |
+
},
|
| 841 |
+
"model.layers.26.self_attn.v_proj": {
|
| 842 |
+
"group_size": 64,
|
| 843 |
+
"bits": 4
|
| 844 |
+
},
|
| 845 |
+
"model.layers.26.self_attn.o_proj": {
|
| 846 |
+
"group_size": 64,
|
| 847 |
+
"bits": 4
|
| 848 |
+
},
|
| 849 |
+
"model.layers.26.self_attn.rope": false,
|
| 850 |
+
"model.layers.26.mlp.gate_proj": {
|
| 851 |
+
"group_size": 64,
|
| 852 |
+
"bits": 4
|
| 853 |
+
},
|
| 854 |
+
"model.layers.26.mlp.down_proj": {
|
| 855 |
+
"group_size": 64,
|
| 856 |
+
"bits": 4
|
| 857 |
+
},
|
| 858 |
+
"model.layers.26.mlp.up_proj": {
|
| 859 |
+
"group_size": 64,
|
| 860 |
+
"bits": 4
|
| 861 |
+
},
|
| 862 |
+
"model.layers.26.input_layernorm": false,
|
| 863 |
+
"model.layers.26.post_attention_layernorm": false,
|
| 864 |
+
"model.layers.27.self_attn.q_proj": {
|
| 865 |
+
"group_size": 64,
|
| 866 |
+
"bits": 4
|
| 867 |
+
},
|
| 868 |
+
"model.layers.27.self_attn.k_proj": {
|
| 869 |
+
"group_size": 64,
|
| 870 |
+
"bits": 4
|
| 871 |
+
},
|
| 872 |
+
"model.layers.27.self_attn.v_proj": {
|
| 873 |
+
"group_size": 64,
|
| 874 |
+
"bits": 4
|
| 875 |
+
},
|
| 876 |
+
"model.layers.27.self_attn.o_proj": {
|
| 877 |
+
"group_size": 64,
|
| 878 |
+
"bits": 4
|
| 879 |
+
},
|
| 880 |
+
"model.layers.27.self_attn.rope": false,
|
| 881 |
+
"model.layers.27.mlp.gate_proj": {
|
| 882 |
+
"group_size": 64,
|
| 883 |
+
"bits": 4
|
| 884 |
+
},
|
| 885 |
+
"model.layers.27.mlp.down_proj": {
|
| 886 |
+
"group_size": 64,
|
| 887 |
+
"bits": 4
|
| 888 |
+
},
|
| 889 |
+
"model.layers.27.mlp.up_proj": {
|
| 890 |
+
"group_size": 64,
|
| 891 |
+
"bits": 4
|
| 892 |
+
},
|
| 893 |
+
"model.layers.27.input_layernorm": false,
|
| 894 |
+
"model.layers.27.post_attention_layernorm": false,
|
| 895 |
+
"model.layers.28.self_attn.q_proj": {
|
| 896 |
+
"group_size": 64,
|
| 897 |
+
"bits": 4
|
| 898 |
+
},
|
| 899 |
+
"model.layers.28.self_attn.k_proj": {
|
| 900 |
+
"group_size": 64,
|
| 901 |
+
"bits": 4
|
| 902 |
+
},
|
| 903 |
+
"model.layers.28.self_attn.v_proj": {
|
| 904 |
+
"group_size": 64,
|
| 905 |
+
"bits": 4
|
| 906 |
+
},
|
| 907 |
+
"model.layers.28.self_attn.o_proj": {
|
| 908 |
+
"group_size": 64,
|
| 909 |
+
"bits": 4
|
| 910 |
+
},
|
| 911 |
+
"model.layers.28.self_attn.rope": false,
|
| 912 |
+
"model.layers.28.mlp.gate_proj": {
|
| 913 |
+
"group_size": 64,
|
| 914 |
+
"bits": 4
|
| 915 |
+
},
|
| 916 |
+
"model.layers.28.mlp.down_proj": {
|
| 917 |
+
"group_size": 64,
|
| 918 |
+
"bits": 4
|
| 919 |
+
},
|
| 920 |
+
"model.layers.28.mlp.up_proj": {
|
| 921 |
+
"group_size": 64,
|
| 922 |
+
"bits": 4
|
| 923 |
+
},
|
| 924 |
+
"model.layers.28.input_layernorm": false,
|
| 925 |
+
"model.layers.28.post_attention_layernorm": false,
|
| 926 |
+
"model.layers.29.self_attn.q_proj": {
|
| 927 |
+
"group_size": 64,
|
| 928 |
+
"bits": 4
|
| 929 |
+
},
|
| 930 |
+
"model.layers.29.self_attn.k_proj": {
|
| 931 |
+
"group_size": 64,
|
| 932 |
+
"bits": 4
|
| 933 |
+
},
|
| 934 |
+
"model.layers.29.self_attn.v_proj": {
|
| 935 |
+
"group_size": 64,
|
| 936 |
+
"bits": 4
|
| 937 |
+
},
|
| 938 |
+
"model.layers.29.self_attn.o_proj": {
|
| 939 |
+
"group_size": 64,
|
| 940 |
+
"bits": 4
|
| 941 |
+
},
|
| 942 |
+
"model.layers.29.self_attn.rope": false,
|
| 943 |
+
"model.layers.29.mlp.gate_proj": {
|
| 944 |
+
"group_size": 64,
|
| 945 |
+
"bits": 4
|
| 946 |
+
},
|
| 947 |
+
"model.layers.29.mlp.down_proj": {
|
| 948 |
+
"group_size": 64,
|
| 949 |
+
"bits": 4
|
| 950 |
+
},
|
| 951 |
+
"model.layers.29.mlp.up_proj": {
|
| 952 |
+
"group_size": 64,
|
| 953 |
+
"bits": 4
|
| 954 |
+
},
|
| 955 |
+
"model.layers.29.input_layernorm": false,
|
| 956 |
+
"model.layers.29.post_attention_layernorm": false,
|
| 957 |
+
"model.layers.30.self_attn.q_proj": {
|
| 958 |
+
"group_size": 64,
|
| 959 |
+
"bits": 4
|
| 960 |
+
},
|
| 961 |
+
"model.layers.30.self_attn.k_proj": {
|
| 962 |
+
"group_size": 64,
|
| 963 |
+
"bits": 4
|
| 964 |
+
},
|
| 965 |
+
"model.layers.30.self_attn.v_proj": {
|
| 966 |
+
"group_size": 64,
|
| 967 |
+
"bits": 4
|
| 968 |
+
},
|
| 969 |
+
"model.layers.30.self_attn.o_proj": {
|
| 970 |
+
"group_size": 64,
|
| 971 |
+
"bits": 4
|
| 972 |
+
},
|
| 973 |
+
"model.layers.30.self_attn.rope": false,
|
| 974 |
+
"model.layers.30.mlp.gate_proj": {
|
| 975 |
+
"group_size": 64,
|
| 976 |
+
"bits": 4
|
| 977 |
+
},
|
| 978 |
+
"model.layers.30.mlp.down_proj": {
|
| 979 |
+
"group_size": 64,
|
| 980 |
+
"bits": 4
|
| 981 |
+
},
|
| 982 |
+
"model.layers.30.mlp.up_proj": {
|
| 983 |
+
"group_size": 64,
|
| 984 |
+
"bits": 4
|
| 985 |
+
},
|
| 986 |
+
"model.layers.30.input_layernorm": false,
|
| 987 |
+
"model.layers.30.post_attention_layernorm": false,
|
| 988 |
+
"model.layers.31.self_attn.q_proj": {
|
| 989 |
+
"group_size": 64,
|
| 990 |
+
"bits": 4
|
| 991 |
+
},
|
| 992 |
+
"model.layers.31.self_attn.k_proj": {
|
| 993 |
+
"group_size": 64,
|
| 994 |
+
"bits": 4
|
| 995 |
+
},
|
| 996 |
+
"model.layers.31.self_attn.v_proj": {
|
| 997 |
+
"group_size": 64,
|
| 998 |
+
"bits": 4
|
| 999 |
+
},
|
| 1000 |
+
"model.layers.31.self_attn.o_proj": {
|
| 1001 |
+
"group_size": 64,
|
| 1002 |
+
"bits": 4
|
| 1003 |
+
},
|
| 1004 |
+
"model.layers.31.self_attn.rope": false,
|
| 1005 |
+
"model.layers.31.mlp.gate_proj": {
|
| 1006 |
+
"group_size": 64,
|
| 1007 |
+
"bits": 4
|
| 1008 |
+
},
|
| 1009 |
+
"model.layers.31.mlp.down_proj": {
|
| 1010 |
+
"group_size": 64,
|
| 1011 |
+
"bits": 4
|
| 1012 |
+
},
|
| 1013 |
+
"model.layers.31.mlp.up_proj": {
|
| 1014 |
+
"group_size": 64,
|
| 1015 |
+
"bits": 4
|
| 1016 |
+
},
|
| 1017 |
+
"model.layers.31.input_layernorm": false,
|
| 1018 |
+
"model.layers.31.post_attention_layernorm": false,
|
| 1019 |
+
"model.norm": false,
|
| 1020 |
+
"lm_head": {
|
| 1021 |
+
"group_size": 32,
|
| 1022 |
+
"bits": 4
|
| 1023 |
+
}
|
| 1024 |
+
},
|
| 1025 |
+
"resid_pdrop": 0.1,
|
| 1026 |
+
"rms_norm_eps": 1e-06,
|
| 1027 |
+
"rope_theta": 500000.0,
|
| 1028 |
+
"tie_word_embeddings": false,
|
| 1029 |
+
"torch_dtype": "bfloat16",
|
| 1030 |
+
"transformers_version": "4.46.2",
|
| 1031 |
+
"use_cache": true,
|
| 1032 |
+
"vocab_size": 155136
|
| 1033 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.46.2"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36d6b7f06693010b4e7b690f47867526b682b78c16ebed2545290d993c939722
|
| 3 |
+
size 4720782232
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,750 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 4720697344
|
| 4 |
+
},
|
| 5 |
+
"weight_map": {
|
| 6 |
+
"lm_head.biases": "model.safetensors",
|
| 7 |
+
"lm_head.scales": "model.safetensors",
|
| 8 |
+
"lm_head.weight": "model.safetensors",
|
| 9 |
+
"model.embed_tokens.biases": "model.safetensors",
|
| 10 |
+
"model.embed_tokens.scales": "model.safetensors",
|
| 11 |
+
"model.embed_tokens.weight": "model.safetensors",
|
| 12 |
+
"model.layers.0.input_layernorm.weight": "model.safetensors",
|
| 13 |
+
"model.layers.0.mlp.down_proj.biases": "model.safetensors",
|
| 14 |
+
"model.layers.0.mlp.down_proj.scales": "model.safetensors",
|
| 15 |
+
"model.layers.0.mlp.down_proj.weight": "model.safetensors",
|
| 16 |
+
"model.layers.0.mlp.gate_proj.biases": "model.safetensors",
|
| 17 |
+
"model.layers.0.mlp.gate_proj.scales": "model.safetensors",
|
| 18 |
+
"model.layers.0.mlp.gate_proj.weight": "model.safetensors",
|
| 19 |
+
"model.layers.0.mlp.up_proj.biases": "model.safetensors",
|
| 20 |
+
"model.layers.0.mlp.up_proj.scales": "model.safetensors",
|
| 21 |
+
"model.layers.0.mlp.up_proj.weight": "model.safetensors",
|
| 22 |
+
"model.layers.0.post_attention_layernorm.weight": "model.safetensors",
|
| 23 |
+
"model.layers.0.self_attn.k_proj.biases": "model.safetensors",
|
| 24 |
+
"model.layers.0.self_attn.k_proj.scales": "model.safetensors",
|
| 25 |
+
"model.layers.0.self_attn.k_proj.weight": "model.safetensors",
|
| 26 |
+
"model.layers.0.self_attn.o_proj.biases": "model.safetensors",
|
| 27 |
+
"model.layers.0.self_attn.o_proj.scales": "model.safetensors",
|
| 28 |
+
"model.layers.0.self_attn.o_proj.weight": "model.safetensors",
|
| 29 |
+
"model.layers.0.self_attn.q_proj.biases": "model.safetensors",
|
| 30 |
+
"model.layers.0.self_attn.q_proj.scales": "model.safetensors",
|
| 31 |
+
"model.layers.0.self_attn.q_proj.weight": "model.safetensors",
|
| 32 |
+
"model.layers.0.self_attn.v_proj.biases": "model.safetensors",
|
| 33 |
+
"model.layers.0.self_attn.v_proj.scales": "model.safetensors",
|
| 34 |
+
"model.layers.0.self_attn.v_proj.weight": "model.safetensors",
|
| 35 |
+
"model.layers.1.input_layernorm.weight": "model.safetensors",
|
| 36 |
+
"model.layers.1.mlp.down_proj.biases": "model.safetensors",
|
| 37 |
+
"model.layers.1.mlp.down_proj.scales": "model.safetensors",
|
| 38 |
+
"model.layers.1.mlp.down_proj.weight": "model.safetensors",
|
| 39 |
+
"model.layers.1.mlp.gate_proj.biases": "model.safetensors",
|
| 40 |
+
"model.layers.1.mlp.gate_proj.scales": "model.safetensors",
|
| 41 |
+
"model.layers.1.mlp.gate_proj.weight": "model.safetensors",
|
| 42 |
+
"model.layers.1.mlp.up_proj.biases": "model.safetensors",
|
| 43 |
+
"model.layers.1.mlp.up_proj.scales": "model.safetensors",
|
| 44 |
+
"model.layers.1.mlp.up_proj.weight": "model.safetensors",
|
| 45 |
+
"model.layers.1.post_attention_layernorm.weight": "model.safetensors",
|
| 46 |
+
"model.layers.1.self_attn.k_proj.biases": "model.safetensors",
|
| 47 |
+
"model.layers.1.self_attn.k_proj.scales": "model.safetensors",
|
| 48 |
+
"model.layers.1.self_attn.k_proj.weight": "model.safetensors",
|
| 49 |
+
"model.layers.1.self_attn.o_proj.biases": "model.safetensors",
|
| 50 |
+
"model.layers.1.self_attn.o_proj.scales": "model.safetensors",
|
| 51 |
+
"model.layers.1.self_attn.o_proj.weight": "model.safetensors",
|
| 52 |
+
"model.layers.1.self_attn.q_proj.biases": "model.safetensors",
|
| 53 |
+
"model.layers.1.self_attn.q_proj.scales": "model.safetensors",
|
| 54 |
+
"model.layers.1.self_attn.q_proj.weight": "model.safetensors",
|
| 55 |
+
"model.layers.1.self_attn.v_proj.biases": "model.safetensors",
|
| 56 |
+
"model.layers.1.self_attn.v_proj.scales": "model.safetensors",
|
| 57 |
+
"model.layers.1.self_attn.v_proj.weight": "model.safetensors",
|
| 58 |
+
"model.layers.10.input_layernorm.weight": "model.safetensors",
|
| 59 |
+
"model.layers.10.mlp.down_proj.biases": "model.safetensors",
|
| 60 |
+
"model.layers.10.mlp.down_proj.scales": "model.safetensors",
|
| 61 |
+
"model.layers.10.mlp.down_proj.weight": "model.safetensors",
|
| 62 |
+
"model.layers.10.mlp.gate_proj.biases": "model.safetensors",
|
| 63 |
+
"model.layers.10.mlp.gate_proj.scales": "model.safetensors",
|
| 64 |
+
"model.layers.10.mlp.gate_proj.weight": "model.safetensors",
|
| 65 |
+
"model.layers.10.mlp.up_proj.biases": "model.safetensors",
|
| 66 |
+
"model.layers.10.mlp.up_proj.scales": "model.safetensors",
|
| 67 |
+
"model.layers.10.mlp.up_proj.weight": "model.safetensors",
|
| 68 |
+
"model.layers.10.post_attention_layernorm.weight": "model.safetensors",
|
| 69 |
+
"model.layers.10.self_attn.k_proj.biases": "model.safetensors",
|
| 70 |
+
"model.layers.10.self_attn.k_proj.scales": "model.safetensors",
|
| 71 |
+
"model.layers.10.self_attn.k_proj.weight": "model.safetensors",
|
| 72 |
+
"model.layers.10.self_attn.o_proj.biases": "model.safetensors",
|
| 73 |
+
"model.layers.10.self_attn.o_proj.scales": "model.safetensors",
|
| 74 |
+
"model.layers.10.self_attn.o_proj.weight": "model.safetensors",
|
| 75 |
+
"model.layers.10.self_attn.q_proj.biases": "model.safetensors",
|
| 76 |
+
"model.layers.10.self_attn.q_proj.scales": "model.safetensors",
|
| 77 |
+
"model.layers.10.self_attn.q_proj.weight": "model.safetensors",
|
| 78 |
+
"model.layers.10.self_attn.v_proj.biases": "model.safetensors",
|
| 79 |
+
"model.layers.10.self_attn.v_proj.scales": "model.safetensors",
|
| 80 |
+
"model.layers.10.self_attn.v_proj.weight": "model.safetensors",
|
| 81 |
+
"model.layers.11.input_layernorm.weight": "model.safetensors",
|
| 82 |
+
"model.layers.11.mlp.down_proj.biases": "model.safetensors",
|
| 83 |
+
"model.layers.11.mlp.down_proj.scales": "model.safetensors",
|
| 84 |
+
"model.layers.11.mlp.down_proj.weight": "model.safetensors",
|
| 85 |
+
"model.layers.11.mlp.gate_proj.biases": "model.safetensors",
|
| 86 |
+
"model.layers.11.mlp.gate_proj.scales": "model.safetensors",
|
| 87 |
+
"model.layers.11.mlp.gate_proj.weight": "model.safetensors",
|
| 88 |
+
"model.layers.11.mlp.up_proj.biases": "model.safetensors",
|
| 89 |
+
"model.layers.11.mlp.up_proj.scales": "model.safetensors",
|
| 90 |
+
"model.layers.11.mlp.up_proj.weight": "model.safetensors",
|
| 91 |
+
"model.layers.11.post_attention_layernorm.weight": "model.safetensors",
|
| 92 |
+
"model.layers.11.self_attn.k_proj.biases": "model.safetensors",
|
| 93 |
+
"model.layers.11.self_attn.k_proj.scales": "model.safetensors",
|
| 94 |
+
"model.layers.11.self_attn.k_proj.weight": "model.safetensors",
|
| 95 |
+
"model.layers.11.self_attn.o_proj.biases": "model.safetensors",
|
| 96 |
+
"model.layers.11.self_attn.o_proj.scales": "model.safetensors",
|
| 97 |
+
"model.layers.11.self_attn.o_proj.weight": "model.safetensors",
|
| 98 |
+
"model.layers.11.self_attn.q_proj.biases": "model.safetensors",
|
| 99 |
+
"model.layers.11.self_attn.q_proj.scales": "model.safetensors",
|
| 100 |
+
"model.layers.11.self_attn.q_proj.weight": "model.safetensors",
|
| 101 |
+
"model.layers.11.self_attn.v_proj.biases": "model.safetensors",
|
| 102 |
+
"model.layers.11.self_attn.v_proj.scales": "model.safetensors",
|
| 103 |
+
"model.layers.11.self_attn.v_proj.weight": "model.safetensors",
|
| 104 |
+
"model.layers.12.input_layernorm.weight": "model.safetensors",
|
| 105 |
+
"model.layers.12.mlp.down_proj.biases": "model.safetensors",
|
| 106 |
+
"model.layers.12.mlp.down_proj.scales": "model.safetensors",
|
| 107 |
+
"model.layers.12.mlp.down_proj.weight": "model.safetensors",
|
| 108 |
+
"model.layers.12.mlp.gate_proj.biases": "model.safetensors",
|
| 109 |
+
"model.layers.12.mlp.gate_proj.scales": "model.safetensors",
|
| 110 |
+
"model.layers.12.mlp.gate_proj.weight": "model.safetensors",
|
| 111 |
+
"model.layers.12.mlp.up_proj.biases": "model.safetensors",
|
| 112 |
+
"model.layers.12.mlp.up_proj.scales": "model.safetensors",
|
| 113 |
+
"model.layers.12.mlp.up_proj.weight": "model.safetensors",
|
| 114 |
+
"model.layers.12.post_attention_layernorm.weight": "model.safetensors",
|
| 115 |
+
"model.layers.12.self_attn.k_proj.biases": "model.safetensors",
|
| 116 |
+
"model.layers.12.self_attn.k_proj.scales": "model.safetensors",
|
| 117 |
+
"model.layers.12.self_attn.k_proj.weight": "model.safetensors",
|
| 118 |
+
"model.layers.12.self_attn.o_proj.biases": "model.safetensors",
|
| 119 |
+
"model.layers.12.self_attn.o_proj.scales": "model.safetensors",
|
| 120 |
+
"model.layers.12.self_attn.o_proj.weight": "model.safetensors",
|
| 121 |
+
"model.layers.12.self_attn.q_proj.biases": "model.safetensors",
|
| 122 |
+
"model.layers.12.self_attn.q_proj.scales": "model.safetensors",
|
| 123 |
+
"model.layers.12.self_attn.q_proj.weight": "model.safetensors",
|
| 124 |
+
"model.layers.12.self_attn.v_proj.biases": "model.safetensors",
|
| 125 |
+
"model.layers.12.self_attn.v_proj.scales": "model.safetensors",
|
| 126 |
+
"model.layers.12.self_attn.v_proj.weight": "model.safetensors",
|
| 127 |
+
"model.layers.13.input_layernorm.weight": "model.safetensors",
|
| 128 |
+
"model.layers.13.mlp.down_proj.biases": "model.safetensors",
|
| 129 |
+
"model.layers.13.mlp.down_proj.scales": "model.safetensors",
|
| 130 |
+
"model.layers.13.mlp.down_proj.weight": "model.safetensors",
|
| 131 |
+
"model.layers.13.mlp.gate_proj.biases": "model.safetensors",
|
| 132 |
+
"model.layers.13.mlp.gate_proj.scales": "model.safetensors",
|
| 133 |
+
"model.layers.13.mlp.gate_proj.weight": "model.safetensors",
|
| 134 |
+
"model.layers.13.mlp.up_proj.biases": "model.safetensors",
|
| 135 |
+
"model.layers.13.mlp.up_proj.scales": "model.safetensors",
|
| 136 |
+
"model.layers.13.mlp.up_proj.weight": "model.safetensors",
|
| 137 |
+
"model.layers.13.post_attention_layernorm.weight": "model.safetensors",
|
| 138 |
+
"model.layers.13.self_attn.k_proj.biases": "model.safetensors",
|
| 139 |
+
"model.layers.13.self_attn.k_proj.scales": "model.safetensors",
|
| 140 |
+
"model.layers.13.self_attn.k_proj.weight": "model.safetensors",
|
| 141 |
+
"model.layers.13.self_attn.o_proj.biases": "model.safetensors",
|
| 142 |
+
"model.layers.13.self_attn.o_proj.scales": "model.safetensors",
|
| 143 |
+
"model.layers.13.self_attn.o_proj.weight": "model.safetensors",
|
| 144 |
+
"model.layers.13.self_attn.q_proj.biases": "model.safetensors",
|
| 145 |
+
"model.layers.13.self_attn.q_proj.scales": "model.safetensors",
|
| 146 |
+
"model.layers.13.self_attn.q_proj.weight": "model.safetensors",
|
| 147 |
+
"model.layers.13.self_attn.v_proj.biases": "model.safetensors",
|
| 148 |
+
"model.layers.13.self_attn.v_proj.scales": "model.safetensors",
|
| 149 |
+
"model.layers.13.self_attn.v_proj.weight": "model.safetensors",
|
| 150 |
+
"model.layers.14.input_layernorm.weight": "model.safetensors",
|
| 151 |
+
"model.layers.14.mlp.down_proj.biases": "model.safetensors",
|
| 152 |
+
"model.layers.14.mlp.down_proj.scales": "model.safetensors",
|
| 153 |
+
"model.layers.14.mlp.down_proj.weight": "model.safetensors",
|
| 154 |
+
"model.layers.14.mlp.gate_proj.biases": "model.safetensors",
|
| 155 |
+
"model.layers.14.mlp.gate_proj.scales": "model.safetensors",
|
| 156 |
+
"model.layers.14.mlp.gate_proj.weight": "model.safetensors",
|
| 157 |
+
"model.layers.14.mlp.up_proj.biases": "model.safetensors",
|
| 158 |
+
"model.layers.14.mlp.up_proj.scales": "model.safetensors",
|
| 159 |
+
"model.layers.14.mlp.up_proj.weight": "model.safetensors",
|
| 160 |
+
"model.layers.14.post_attention_layernorm.weight": "model.safetensors",
|
| 161 |
+
"model.layers.14.self_attn.k_proj.biases": "model.safetensors",
|
| 162 |
+
"model.layers.14.self_attn.k_proj.scales": "model.safetensors",
|
| 163 |
+
"model.layers.14.self_attn.k_proj.weight": "model.safetensors",
|
| 164 |
+
"model.layers.14.self_attn.o_proj.biases": "model.safetensors",
|
| 165 |
+
"model.layers.14.self_attn.o_proj.scales": "model.safetensors",
|
| 166 |
+
"model.layers.14.self_attn.o_proj.weight": "model.safetensors",
|
| 167 |
+
"model.layers.14.self_attn.q_proj.biases": "model.safetensors",
|
| 168 |
+
"model.layers.14.self_attn.q_proj.scales": "model.safetensors",
|
| 169 |
+
"model.layers.14.self_attn.q_proj.weight": "model.safetensors",
|
| 170 |
+
"model.layers.14.self_attn.v_proj.biases": "model.safetensors",
|
| 171 |
+
"model.layers.14.self_attn.v_proj.scales": "model.safetensors",
|
| 172 |
+
"model.layers.14.self_attn.v_proj.weight": "model.safetensors",
|
| 173 |
+
"model.layers.15.input_layernorm.weight": "model.safetensors",
|
| 174 |
+
"model.layers.15.mlp.down_proj.biases": "model.safetensors",
|
| 175 |
+
"model.layers.15.mlp.down_proj.scales": "model.safetensors",
|
| 176 |
+
"model.layers.15.mlp.down_proj.weight": "model.safetensors",
|
| 177 |
+
"model.layers.15.mlp.gate_proj.biases": "model.safetensors",
|
| 178 |
+
"model.layers.15.mlp.gate_proj.scales": "model.safetensors",
|
| 179 |
+
"model.layers.15.mlp.gate_proj.weight": "model.safetensors",
|
| 180 |
+
"model.layers.15.mlp.up_proj.biases": "model.safetensors",
|
| 181 |
+
"model.layers.15.mlp.up_proj.scales": "model.safetensors",
|
| 182 |
+
"model.layers.15.mlp.up_proj.weight": "model.safetensors",
|
| 183 |
+
"model.layers.15.post_attention_layernorm.weight": "model.safetensors",
|
| 184 |
+
"model.layers.15.self_attn.k_proj.biases": "model.safetensors",
|
| 185 |
+
"model.layers.15.self_attn.k_proj.scales": "model.safetensors",
|
| 186 |
+
"model.layers.15.self_attn.k_proj.weight": "model.safetensors",
|
| 187 |
+
"model.layers.15.self_attn.o_proj.biases": "model.safetensors",
|
| 188 |
+
"model.layers.15.self_attn.o_proj.scales": "model.safetensors",
|
| 189 |
+
"model.layers.15.self_attn.o_proj.weight": "model.safetensors",
|
| 190 |
+
"model.layers.15.self_attn.q_proj.biases": "model.safetensors",
|
| 191 |
+
"model.layers.15.self_attn.q_proj.scales": "model.safetensors",
|
| 192 |
+
"model.layers.15.self_attn.q_proj.weight": "model.safetensors",
|
| 193 |
+
"model.layers.15.self_attn.v_proj.biases": "model.safetensors",
|
| 194 |
+
"model.layers.15.self_attn.v_proj.scales": "model.safetensors",
|
| 195 |
+
"model.layers.15.self_attn.v_proj.weight": "model.safetensors",
|
| 196 |
+
"model.layers.16.input_layernorm.weight": "model.safetensors",
|
| 197 |
+
"model.layers.16.mlp.down_proj.biases": "model.safetensors",
|
| 198 |
+
"model.layers.16.mlp.down_proj.scales": "model.safetensors",
|
| 199 |
+
"model.layers.16.mlp.down_proj.weight": "model.safetensors",
|
| 200 |
+
"model.layers.16.mlp.gate_proj.biases": "model.safetensors",
|
| 201 |
+
"model.layers.16.mlp.gate_proj.scales": "model.safetensors",
|
| 202 |
+
"model.layers.16.mlp.gate_proj.weight": "model.safetensors",
|
| 203 |
+
"model.layers.16.mlp.up_proj.biases": "model.safetensors",
|
| 204 |
+
"model.layers.16.mlp.up_proj.scales": "model.safetensors",
|
| 205 |
+
"model.layers.16.mlp.up_proj.weight": "model.safetensors",
|
| 206 |
+
"model.layers.16.post_attention_layernorm.weight": "model.safetensors",
|
| 207 |
+
"model.layers.16.self_attn.k_proj.biases": "model.safetensors",
|
| 208 |
+
"model.layers.16.self_attn.k_proj.scales": "model.safetensors",
|
| 209 |
+
"model.layers.16.self_attn.k_proj.weight": "model.safetensors",
|
| 210 |
+
"model.layers.16.self_attn.o_proj.biases": "model.safetensors",
|
| 211 |
+
"model.layers.16.self_attn.o_proj.scales": "model.safetensors",
|
| 212 |
+
"model.layers.16.self_attn.o_proj.weight": "model.safetensors",
|
| 213 |
+
"model.layers.16.self_attn.q_proj.biases": "model.safetensors",
|
| 214 |
+
"model.layers.16.self_attn.q_proj.scales": "model.safetensors",
|
| 215 |
+
"model.layers.16.self_attn.q_proj.weight": "model.safetensors",
|
| 216 |
+
"model.layers.16.self_attn.v_proj.biases": "model.safetensors",
|
| 217 |
+
"model.layers.16.self_attn.v_proj.scales": "model.safetensors",
|
| 218 |
+
"model.layers.16.self_attn.v_proj.weight": "model.safetensors",
|
| 219 |
+
"model.layers.17.input_layernorm.weight": "model.safetensors",
|
| 220 |
+
"model.layers.17.mlp.down_proj.biases": "model.safetensors",
|
| 221 |
+
"model.layers.17.mlp.down_proj.scales": "model.safetensors",
|
| 222 |
+
"model.layers.17.mlp.down_proj.weight": "model.safetensors",
|
| 223 |
+
"model.layers.17.mlp.gate_proj.biases": "model.safetensors",
|
| 224 |
+
"model.layers.17.mlp.gate_proj.scales": "model.safetensors",
|
| 225 |
+
"model.layers.17.mlp.gate_proj.weight": "model.safetensors",
|
| 226 |
+
"model.layers.17.mlp.up_proj.biases": "model.safetensors",
|
| 227 |
+
"model.layers.17.mlp.up_proj.scales": "model.safetensors",
|
| 228 |
+
"model.layers.17.mlp.up_proj.weight": "model.safetensors",
|
| 229 |
+
"model.layers.17.post_attention_layernorm.weight": "model.safetensors",
|
| 230 |
+
"model.layers.17.self_attn.k_proj.biases": "model.safetensors",
|
| 231 |
+
"model.layers.17.self_attn.k_proj.scales": "model.safetensors",
|
| 232 |
+
"model.layers.17.self_attn.k_proj.weight": "model.safetensors",
|
| 233 |
+
"model.layers.17.self_attn.o_proj.biases": "model.safetensors",
|
| 234 |
+
"model.layers.17.self_attn.o_proj.scales": "model.safetensors",
|
| 235 |
+
"model.layers.17.self_attn.o_proj.weight": "model.safetensors",
|
| 236 |
+
"model.layers.17.self_attn.q_proj.biases": "model.safetensors",
|
| 237 |
+
"model.layers.17.self_attn.q_proj.scales": "model.safetensors",
|
| 238 |
+
"model.layers.17.self_attn.q_proj.weight": "model.safetensors",
|
| 239 |
+
"model.layers.17.self_attn.v_proj.biases": "model.safetensors",
|
| 240 |
+
"model.layers.17.self_attn.v_proj.scales": "model.safetensors",
|
| 241 |
+
"model.layers.17.self_attn.v_proj.weight": "model.safetensors",
|
| 242 |
+
"model.layers.18.input_layernorm.weight": "model.safetensors",
|
| 243 |
+
"model.layers.18.mlp.down_proj.biases": "model.safetensors",
|
| 244 |
+
"model.layers.18.mlp.down_proj.scales": "model.safetensors",
|
| 245 |
+
"model.layers.18.mlp.down_proj.weight": "model.safetensors",
|
| 246 |
+
"model.layers.18.mlp.gate_proj.biases": "model.safetensors",
|
| 247 |
+
"model.layers.18.mlp.gate_proj.scales": "model.safetensors",
|
| 248 |
+
"model.layers.18.mlp.gate_proj.weight": "model.safetensors",
|
| 249 |
+
"model.layers.18.mlp.up_proj.biases": "model.safetensors",
|
| 250 |
+
"model.layers.18.mlp.up_proj.scales": "model.safetensors",
|
| 251 |
+
"model.layers.18.mlp.up_proj.weight": "model.safetensors",
|
| 252 |
+
"model.layers.18.post_attention_layernorm.weight": "model.safetensors",
|
| 253 |
+
"model.layers.18.self_attn.k_proj.biases": "model.safetensors",
|
| 254 |
+
"model.layers.18.self_attn.k_proj.scales": "model.safetensors",
|
| 255 |
+
"model.layers.18.self_attn.k_proj.weight": "model.safetensors",
|
| 256 |
+
"model.layers.18.self_attn.o_proj.biases": "model.safetensors",
|
| 257 |
+
"model.layers.18.self_attn.o_proj.scales": "model.safetensors",
|
| 258 |
+
"model.layers.18.self_attn.o_proj.weight": "model.safetensors",
|
| 259 |
+
"model.layers.18.self_attn.q_proj.biases": "model.safetensors",
|
| 260 |
+
"model.layers.18.self_attn.q_proj.scales": "model.safetensors",
|
| 261 |
+
"model.layers.18.self_attn.q_proj.weight": "model.safetensors",
|
| 262 |
+
"model.layers.18.self_attn.v_proj.biases": "model.safetensors",
|
| 263 |
+
"model.layers.18.self_attn.v_proj.scales": "model.safetensors",
|
| 264 |
+
"model.layers.18.self_attn.v_proj.weight": "model.safetensors",
|
| 265 |
+
"model.layers.19.input_layernorm.weight": "model.safetensors",
|
| 266 |
+
"model.layers.19.mlp.down_proj.biases": "model.safetensors",
|
| 267 |
+
"model.layers.19.mlp.down_proj.scales": "model.safetensors",
|
| 268 |
+
"model.layers.19.mlp.down_proj.weight": "model.safetensors",
|
| 269 |
+
"model.layers.19.mlp.gate_proj.biases": "model.safetensors",
|
| 270 |
+
"model.layers.19.mlp.gate_proj.scales": "model.safetensors",
|
| 271 |
+
"model.layers.19.mlp.gate_proj.weight": "model.safetensors",
|
| 272 |
+
"model.layers.19.mlp.up_proj.biases": "model.safetensors",
|
| 273 |
+
"model.layers.19.mlp.up_proj.scales": "model.safetensors",
|
| 274 |
+
"model.layers.19.mlp.up_proj.weight": "model.safetensors",
|
| 275 |
+
"model.layers.19.post_attention_layernorm.weight": "model.safetensors",
|
| 276 |
+
"model.layers.19.self_attn.k_proj.biases": "model.safetensors",
|
| 277 |
+
"model.layers.19.self_attn.k_proj.scales": "model.safetensors",
|
| 278 |
+
"model.layers.19.self_attn.k_proj.weight": "model.safetensors",
|
| 279 |
+
"model.layers.19.self_attn.o_proj.biases": "model.safetensors",
|
| 280 |
+
"model.layers.19.self_attn.o_proj.scales": "model.safetensors",
|
| 281 |
+
"model.layers.19.self_attn.o_proj.weight": "model.safetensors",
|
| 282 |
+
"model.layers.19.self_attn.q_proj.biases": "model.safetensors",
|
| 283 |
+
"model.layers.19.self_attn.q_proj.scales": "model.safetensors",
|
| 284 |
+
"model.layers.19.self_attn.q_proj.weight": "model.safetensors",
|
| 285 |
+
"model.layers.19.self_attn.v_proj.biases": "model.safetensors",
|
| 286 |
+
"model.layers.19.self_attn.v_proj.scales": "model.safetensors",
|
| 287 |
+
"model.layers.19.self_attn.v_proj.weight": "model.safetensors",
|
| 288 |
+
"model.layers.2.input_layernorm.weight": "model.safetensors",
|
| 289 |
+
"model.layers.2.mlp.down_proj.biases": "model.safetensors",
|
| 290 |
+
"model.layers.2.mlp.down_proj.scales": "model.safetensors",
|
| 291 |
+
"model.layers.2.mlp.down_proj.weight": "model.safetensors",
|
| 292 |
+
"model.layers.2.mlp.gate_proj.biases": "model.safetensors",
|
| 293 |
+
"model.layers.2.mlp.gate_proj.scales": "model.safetensors",
|
| 294 |
+
"model.layers.2.mlp.gate_proj.weight": "model.safetensors",
|
| 295 |
+
"model.layers.2.mlp.up_proj.biases": "model.safetensors",
|
| 296 |
+
"model.layers.2.mlp.up_proj.scales": "model.safetensors",
|
| 297 |
+
"model.layers.2.mlp.up_proj.weight": "model.safetensors",
|
| 298 |
+
"model.layers.2.post_attention_layernorm.weight": "model.safetensors",
|
| 299 |
+
"model.layers.2.self_attn.k_proj.biases": "model.safetensors",
|
| 300 |
+
"model.layers.2.self_attn.k_proj.scales": "model.safetensors",
|
| 301 |
+
"model.layers.2.self_attn.k_proj.weight": "model.safetensors",
|
| 302 |
+
"model.layers.2.self_attn.o_proj.biases": "model.safetensors",
|
| 303 |
+
"model.layers.2.self_attn.o_proj.scales": "model.safetensors",
|
| 304 |
+
"model.layers.2.self_attn.o_proj.weight": "model.safetensors",
|
| 305 |
+
"model.layers.2.self_attn.q_proj.biases": "model.safetensors",
|
| 306 |
+
"model.layers.2.self_attn.q_proj.scales": "model.safetensors",
|
| 307 |
+
"model.layers.2.self_attn.q_proj.weight": "model.safetensors",
|
| 308 |
+
"model.layers.2.self_attn.v_proj.biases": "model.safetensors",
|
| 309 |
+
"model.layers.2.self_attn.v_proj.scales": "model.safetensors",
|
| 310 |
+
"model.layers.2.self_attn.v_proj.weight": "model.safetensors",
|
| 311 |
+
"model.layers.20.input_layernorm.weight": "model.safetensors",
|
| 312 |
+
"model.layers.20.mlp.down_proj.biases": "model.safetensors",
|
| 313 |
+
"model.layers.20.mlp.down_proj.scales": "model.safetensors",
|
| 314 |
+
"model.layers.20.mlp.down_proj.weight": "model.safetensors",
|
| 315 |
+
"model.layers.20.mlp.gate_proj.biases": "model.safetensors",
|
| 316 |
+
"model.layers.20.mlp.gate_proj.scales": "model.safetensors",
|
| 317 |
+
"model.layers.20.mlp.gate_proj.weight": "model.safetensors",
|
| 318 |
+
"model.layers.20.mlp.up_proj.biases": "model.safetensors",
|
| 319 |
+
"model.layers.20.mlp.up_proj.scales": "model.safetensors",
|
| 320 |
+
"model.layers.20.mlp.up_proj.weight": "model.safetensors",
|
| 321 |
+
"model.layers.20.post_attention_layernorm.weight": "model.safetensors",
|
| 322 |
+
"model.layers.20.self_attn.k_proj.biases": "model.safetensors",
|
| 323 |
+
"model.layers.20.self_attn.k_proj.scales": "model.safetensors",
|
| 324 |
+
"model.layers.20.self_attn.k_proj.weight": "model.safetensors",
|
| 325 |
+
"model.layers.20.self_attn.o_proj.biases": "model.safetensors",
|
| 326 |
+
"model.layers.20.self_attn.o_proj.scales": "model.safetensors",
|
| 327 |
+
"model.layers.20.self_attn.o_proj.weight": "model.safetensors",
|
| 328 |
+
"model.layers.20.self_attn.q_proj.biases": "model.safetensors",
|
| 329 |
+
"model.layers.20.self_attn.q_proj.scales": "model.safetensors",
|
| 330 |
+
"model.layers.20.self_attn.q_proj.weight": "model.safetensors",
|
| 331 |
+
"model.layers.20.self_attn.v_proj.biases": "model.safetensors",
|
| 332 |
+
"model.layers.20.self_attn.v_proj.scales": "model.safetensors",
|
| 333 |
+
"model.layers.20.self_attn.v_proj.weight": "model.safetensors",
|
| 334 |
+
"model.layers.21.input_layernorm.weight": "model.safetensors",
|
| 335 |
+
"model.layers.21.mlp.down_proj.biases": "model.safetensors",
|
| 336 |
+
"model.layers.21.mlp.down_proj.scales": "model.safetensors",
|
| 337 |
+
"model.layers.21.mlp.down_proj.weight": "model.safetensors",
|
| 338 |
+
"model.layers.21.mlp.gate_proj.biases": "model.safetensors",
|
| 339 |
+
"model.layers.21.mlp.gate_proj.scales": "model.safetensors",
|
| 340 |
+
"model.layers.21.mlp.gate_proj.weight": "model.safetensors",
|
| 341 |
+
"model.layers.21.mlp.up_proj.biases": "model.safetensors",
|
| 342 |
+
"model.layers.21.mlp.up_proj.scales": "model.safetensors",
|
| 343 |
+
"model.layers.21.mlp.up_proj.weight": "model.safetensors",
|
| 344 |
+
"model.layers.21.post_attention_layernorm.weight": "model.safetensors",
|
| 345 |
+
"model.layers.21.self_attn.k_proj.biases": "model.safetensors",
|
| 346 |
+
"model.layers.21.self_attn.k_proj.scales": "model.safetensors",
|
| 347 |
+
"model.layers.21.self_attn.k_proj.weight": "model.safetensors",
|
| 348 |
+
"model.layers.21.self_attn.o_proj.biases": "model.safetensors",
|
| 349 |
+
"model.layers.21.self_attn.o_proj.scales": "model.safetensors",
|
| 350 |
+
"model.layers.21.self_attn.o_proj.weight": "model.safetensors",
|
| 351 |
+
"model.layers.21.self_attn.q_proj.biases": "model.safetensors",
|
| 352 |
+
"model.layers.21.self_attn.q_proj.scales": "model.safetensors",
|
| 353 |
+
"model.layers.21.self_attn.q_proj.weight": "model.safetensors",
|
| 354 |
+
"model.layers.21.self_attn.v_proj.biases": "model.safetensors",
|
| 355 |
+
"model.layers.21.self_attn.v_proj.scales": "model.safetensors",
|
| 356 |
+
"model.layers.21.self_attn.v_proj.weight": "model.safetensors",
|
| 357 |
+
"model.layers.22.input_layernorm.weight": "model.safetensors",
|
| 358 |
+
"model.layers.22.mlp.down_proj.biases": "model.safetensors",
|
| 359 |
+
"model.layers.22.mlp.down_proj.scales": "model.safetensors",
|
| 360 |
+
"model.layers.22.mlp.down_proj.weight": "model.safetensors",
|
| 361 |
+
"model.layers.22.mlp.gate_proj.biases": "model.safetensors",
|
| 362 |
+
"model.layers.22.mlp.gate_proj.scales": "model.safetensors",
|
| 363 |
+
"model.layers.22.mlp.gate_proj.weight": "model.safetensors",
|
| 364 |
+
"model.layers.22.mlp.up_proj.biases": "model.safetensors",
|
| 365 |
+
"model.layers.22.mlp.up_proj.scales": "model.safetensors",
|
| 366 |
+
"model.layers.22.mlp.up_proj.weight": "model.safetensors",
|
| 367 |
+
"model.layers.22.post_attention_layernorm.weight": "model.safetensors",
|
| 368 |
+
"model.layers.22.self_attn.k_proj.biases": "model.safetensors",
|
| 369 |
+
"model.layers.22.self_attn.k_proj.scales": "model.safetensors",
|
| 370 |
+
"model.layers.22.self_attn.k_proj.weight": "model.safetensors",
|
| 371 |
+
"model.layers.22.self_attn.o_proj.biases": "model.safetensors",
|
| 372 |
+
"model.layers.22.self_attn.o_proj.scales": "model.safetensors",
|
| 373 |
+
"model.layers.22.self_attn.o_proj.weight": "model.safetensors",
|
| 374 |
+
"model.layers.22.self_attn.q_proj.biases": "model.safetensors",
|
| 375 |
+
"model.layers.22.self_attn.q_proj.scales": "model.safetensors",
|
| 376 |
+
"model.layers.22.self_attn.q_proj.weight": "model.safetensors",
|
| 377 |
+
"model.layers.22.self_attn.v_proj.biases": "model.safetensors",
|
| 378 |
+
"model.layers.22.self_attn.v_proj.scales": "model.safetensors",
|
| 379 |
+
"model.layers.22.self_attn.v_proj.weight": "model.safetensors",
|
| 380 |
+
"model.layers.23.input_layernorm.weight": "model.safetensors",
|
| 381 |
+
"model.layers.23.mlp.down_proj.biases": "model.safetensors",
|
| 382 |
+
"model.layers.23.mlp.down_proj.scales": "model.safetensors",
|
| 383 |
+
"model.layers.23.mlp.down_proj.weight": "model.safetensors",
|
| 384 |
+
"model.layers.23.mlp.gate_proj.biases": "model.safetensors",
|
| 385 |
+
"model.layers.23.mlp.gate_proj.scales": "model.safetensors",
|
| 386 |
+
"model.layers.23.mlp.gate_proj.weight": "model.safetensors",
|
| 387 |
+
"model.layers.23.mlp.up_proj.biases": "model.safetensors",
|
| 388 |
+
"model.layers.23.mlp.up_proj.scales": "model.safetensors",
|
| 389 |
+
"model.layers.23.mlp.up_proj.weight": "model.safetensors",
|
| 390 |
+
"model.layers.23.post_attention_layernorm.weight": "model.safetensors",
|
| 391 |
+
"model.layers.23.self_attn.k_proj.biases": "model.safetensors",
|
| 392 |
+
"model.layers.23.self_attn.k_proj.scales": "model.safetensors",
|
| 393 |
+
"model.layers.23.self_attn.k_proj.weight": "model.safetensors",
|
| 394 |
+
"model.layers.23.self_attn.o_proj.biases": "model.safetensors",
|
| 395 |
+
"model.layers.23.self_attn.o_proj.scales": "model.safetensors",
|
| 396 |
+
"model.layers.23.self_attn.o_proj.weight": "model.safetensors",
|
| 397 |
+
"model.layers.23.self_attn.q_proj.biases": "model.safetensors",
|
| 398 |
+
"model.layers.23.self_attn.q_proj.scales": "model.safetensors",
|
| 399 |
+
"model.layers.23.self_attn.q_proj.weight": "model.safetensors",
|
| 400 |
+
"model.layers.23.self_attn.v_proj.biases": "model.safetensors",
|
| 401 |
+
"model.layers.23.self_attn.v_proj.scales": "model.safetensors",
|
| 402 |
+
"model.layers.23.self_attn.v_proj.weight": "model.safetensors",
|
| 403 |
+
"model.layers.24.input_layernorm.weight": "model.safetensors",
|
| 404 |
+
"model.layers.24.mlp.down_proj.biases": "model.safetensors",
|
| 405 |
+
"model.layers.24.mlp.down_proj.scales": "model.safetensors",
|
| 406 |
+
"model.layers.24.mlp.down_proj.weight": "model.safetensors",
|
| 407 |
+
"model.layers.24.mlp.gate_proj.biases": "model.safetensors",
|
| 408 |
+
"model.layers.24.mlp.gate_proj.scales": "model.safetensors",
|
| 409 |
+
"model.layers.24.mlp.gate_proj.weight": "model.safetensors",
|
| 410 |
+
"model.layers.24.mlp.up_proj.biases": "model.safetensors",
|
| 411 |
+
"model.layers.24.mlp.up_proj.scales": "model.safetensors",
|
| 412 |
+
"model.layers.24.mlp.up_proj.weight": "model.safetensors",
|
| 413 |
+
"model.layers.24.post_attention_layernorm.weight": "model.safetensors",
|
| 414 |
+
"model.layers.24.self_attn.k_proj.biases": "model.safetensors",
|
| 415 |
+
"model.layers.24.self_attn.k_proj.scales": "model.safetensors",
|
| 416 |
+
"model.layers.24.self_attn.k_proj.weight": "model.safetensors",
|
| 417 |
+
"model.layers.24.self_attn.o_proj.biases": "model.safetensors",
|
| 418 |
+
"model.layers.24.self_attn.o_proj.scales": "model.safetensors",
|
| 419 |
+
"model.layers.24.self_attn.o_proj.weight": "model.safetensors",
|
| 420 |
+
"model.layers.24.self_attn.q_proj.biases": "model.safetensors",
|
| 421 |
+
"model.layers.24.self_attn.q_proj.scales": "model.safetensors",
|
| 422 |
+
"model.layers.24.self_attn.q_proj.weight": "model.safetensors",
|
| 423 |
+
"model.layers.24.self_attn.v_proj.biases": "model.safetensors",
|
| 424 |
+
"model.layers.24.self_attn.v_proj.scales": "model.safetensors",
|
| 425 |
+
"model.layers.24.self_attn.v_proj.weight": "model.safetensors",
|
| 426 |
+
"model.layers.25.input_layernorm.weight": "model.safetensors",
|
| 427 |
+
"model.layers.25.mlp.down_proj.biases": "model.safetensors",
|
| 428 |
+
"model.layers.25.mlp.down_proj.scales": "model.safetensors",
|
| 429 |
+
"model.layers.25.mlp.down_proj.weight": "model.safetensors",
|
| 430 |
+
"model.layers.25.mlp.gate_proj.biases": "model.safetensors",
|
| 431 |
+
"model.layers.25.mlp.gate_proj.scales": "model.safetensors",
|
| 432 |
+
"model.layers.25.mlp.gate_proj.weight": "model.safetensors",
|
| 433 |
+
"model.layers.25.mlp.up_proj.biases": "model.safetensors",
|
| 434 |
+
"model.layers.25.mlp.up_proj.scales": "model.safetensors",
|
| 435 |
+
"model.layers.25.mlp.up_proj.weight": "model.safetensors",
|
| 436 |
+
"model.layers.25.post_attention_layernorm.weight": "model.safetensors",
|
| 437 |
+
"model.layers.25.self_attn.k_proj.biases": "model.safetensors",
|
| 438 |
+
"model.layers.25.self_attn.k_proj.scales": "model.safetensors",
|
| 439 |
+
"model.layers.25.self_attn.k_proj.weight": "model.safetensors",
|
| 440 |
+
"model.layers.25.self_attn.o_proj.biases": "model.safetensors",
|
| 441 |
+
"model.layers.25.self_attn.o_proj.scales": "model.safetensors",
|
| 442 |
+
"model.layers.25.self_attn.o_proj.weight": "model.safetensors",
|
| 443 |
+
"model.layers.25.self_attn.q_proj.biases": "model.safetensors",
|
| 444 |
+
"model.layers.25.self_attn.q_proj.scales": "model.safetensors",
|
| 445 |
+
"model.layers.25.self_attn.q_proj.weight": "model.safetensors",
|
| 446 |
+
"model.layers.25.self_attn.v_proj.biases": "model.safetensors",
|
| 447 |
+
"model.layers.25.self_attn.v_proj.scales": "model.safetensors",
|
| 448 |
+
"model.layers.25.self_attn.v_proj.weight": "model.safetensors",
|
| 449 |
+
"model.layers.26.input_layernorm.weight": "model.safetensors",
|
| 450 |
+
"model.layers.26.mlp.down_proj.biases": "model.safetensors",
|
| 451 |
+
"model.layers.26.mlp.down_proj.scales": "model.safetensors",
|
| 452 |
+
"model.layers.26.mlp.down_proj.weight": "model.safetensors",
|
| 453 |
+
"model.layers.26.mlp.gate_proj.biases": "model.safetensors",
|
| 454 |
+
"model.layers.26.mlp.gate_proj.scales": "model.safetensors",
|
| 455 |
+
"model.layers.26.mlp.gate_proj.weight": "model.safetensors",
|
| 456 |
+
"model.layers.26.mlp.up_proj.biases": "model.safetensors",
|
| 457 |
+
"model.layers.26.mlp.up_proj.scales": "model.safetensors",
|
| 458 |
+
"model.layers.26.mlp.up_proj.weight": "model.safetensors",
|
| 459 |
+
"model.layers.26.post_attention_layernorm.weight": "model.safetensors",
|
| 460 |
+
"model.layers.26.self_attn.k_proj.biases": "model.safetensors",
|
| 461 |
+
"model.layers.26.self_attn.k_proj.scales": "model.safetensors",
|
| 462 |
+
"model.layers.26.self_attn.k_proj.weight": "model.safetensors",
|
| 463 |
+
"model.layers.26.self_attn.o_proj.biases": "model.safetensors",
|
| 464 |
+
"model.layers.26.self_attn.o_proj.scales": "model.safetensors",
|
| 465 |
+
"model.layers.26.self_attn.o_proj.weight": "model.safetensors",
|
| 466 |
+
"model.layers.26.self_attn.q_proj.biases": "model.safetensors",
|
| 467 |
+
"model.layers.26.self_attn.q_proj.scales": "model.safetensors",
|
| 468 |
+
"model.layers.26.self_attn.q_proj.weight": "model.safetensors",
|
| 469 |
+
"model.layers.26.self_attn.v_proj.biases": "model.safetensors",
|
| 470 |
+
"model.layers.26.self_attn.v_proj.scales": "model.safetensors",
|
| 471 |
+
"model.layers.26.self_attn.v_proj.weight": "model.safetensors",
|
| 472 |
+
"model.layers.27.input_layernorm.weight": "model.safetensors",
|
| 473 |
+
"model.layers.27.mlp.down_proj.biases": "model.safetensors",
|
| 474 |
+
"model.layers.27.mlp.down_proj.scales": "model.safetensors",
|
| 475 |
+
"model.layers.27.mlp.down_proj.weight": "model.safetensors",
|
| 476 |
+
"model.layers.27.mlp.gate_proj.biases": "model.safetensors",
|
| 477 |
+
"model.layers.27.mlp.gate_proj.scales": "model.safetensors",
|
| 478 |
+
"model.layers.27.mlp.gate_proj.weight": "model.safetensors",
|
| 479 |
+
"model.layers.27.mlp.up_proj.biases": "model.safetensors",
|
| 480 |
+
"model.layers.27.mlp.up_proj.scales": "model.safetensors",
|
| 481 |
+
"model.layers.27.mlp.up_proj.weight": "model.safetensors",
|
| 482 |
+
"model.layers.27.post_attention_layernorm.weight": "model.safetensors",
|
| 483 |
+
"model.layers.27.self_attn.k_proj.biases": "model.safetensors",
|
| 484 |
+
"model.layers.27.self_attn.k_proj.scales": "model.safetensors",
|
| 485 |
+
"model.layers.27.self_attn.k_proj.weight": "model.safetensors",
|
| 486 |
+
"model.layers.27.self_attn.o_proj.biases": "model.safetensors",
|
| 487 |
+
"model.layers.27.self_attn.o_proj.scales": "model.safetensors",
|
| 488 |
+
"model.layers.27.self_attn.o_proj.weight": "model.safetensors",
|
| 489 |
+
"model.layers.27.self_attn.q_proj.biases": "model.safetensors",
|
| 490 |
+
"model.layers.27.self_attn.q_proj.scales": "model.safetensors",
|
| 491 |
+
"model.layers.27.self_attn.q_proj.weight": "model.safetensors",
|
| 492 |
+
"model.layers.27.self_attn.v_proj.biases": "model.safetensors",
|
| 493 |
+
"model.layers.27.self_attn.v_proj.scales": "model.safetensors",
|
| 494 |
+
"model.layers.27.self_attn.v_proj.weight": "model.safetensors",
|
| 495 |
+
"model.layers.28.input_layernorm.weight": "model.safetensors",
|
| 496 |
+
"model.layers.28.mlp.down_proj.biases": "model.safetensors",
|
| 497 |
+
"model.layers.28.mlp.down_proj.scales": "model.safetensors",
|
| 498 |
+
"model.layers.28.mlp.down_proj.weight": "model.safetensors",
|
| 499 |
+
"model.layers.28.mlp.gate_proj.biases": "model.safetensors",
|
| 500 |
+
"model.layers.28.mlp.gate_proj.scales": "model.safetensors",
|
| 501 |
+
"model.layers.28.mlp.gate_proj.weight": "model.safetensors",
|
| 502 |
+
"model.layers.28.mlp.up_proj.biases": "model.safetensors",
|
| 503 |
+
"model.layers.28.mlp.up_proj.scales": "model.safetensors",
|
| 504 |
+
"model.layers.28.mlp.up_proj.weight": "model.safetensors",
|
| 505 |
+
"model.layers.28.post_attention_layernorm.weight": "model.safetensors",
|
| 506 |
+
"model.layers.28.self_attn.k_proj.biases": "model.safetensors",
|
| 507 |
+
"model.layers.28.self_attn.k_proj.scales": "model.safetensors",
|
| 508 |
+
"model.layers.28.self_attn.k_proj.weight": "model.safetensors",
|
| 509 |
+
"model.layers.28.self_attn.o_proj.biases": "model.safetensors",
|
| 510 |
+
"model.layers.28.self_attn.o_proj.scales": "model.safetensors",
|
| 511 |
+
"model.layers.28.self_attn.o_proj.weight": "model.safetensors",
|
| 512 |
+
"model.layers.28.self_attn.q_proj.biases": "model.safetensors",
|
| 513 |
+
"model.layers.28.self_attn.q_proj.scales": "model.safetensors",
|
| 514 |
+
"model.layers.28.self_attn.q_proj.weight": "model.safetensors",
|
| 515 |
+
"model.layers.28.self_attn.v_proj.biases": "model.safetensors",
|
| 516 |
+
"model.layers.28.self_attn.v_proj.scales": "model.safetensors",
|
| 517 |
+
"model.layers.28.self_attn.v_proj.weight": "model.safetensors",
|
| 518 |
+
"model.layers.29.input_layernorm.weight": "model.safetensors",
|
| 519 |
+
"model.layers.29.mlp.down_proj.biases": "model.safetensors",
|
| 520 |
+
"model.layers.29.mlp.down_proj.scales": "model.safetensors",
|
| 521 |
+
"model.layers.29.mlp.down_proj.weight": "model.safetensors",
|
| 522 |
+
"model.layers.29.mlp.gate_proj.biases": "model.safetensors",
|
| 523 |
+
"model.layers.29.mlp.gate_proj.scales": "model.safetensors",
|
| 524 |
+
"model.layers.29.mlp.gate_proj.weight": "model.safetensors",
|
| 525 |
+
"model.layers.29.mlp.up_proj.biases": "model.safetensors",
|
| 526 |
+
"model.layers.29.mlp.up_proj.scales": "model.safetensors",
|
| 527 |
+
"model.layers.29.mlp.up_proj.weight": "model.safetensors",
|
| 528 |
+
"model.layers.29.post_attention_layernorm.weight": "model.safetensors",
|
| 529 |
+
"model.layers.29.self_attn.k_proj.biases": "model.safetensors",
|
| 530 |
+
"model.layers.29.self_attn.k_proj.scales": "model.safetensors",
|
| 531 |
+
"model.layers.29.self_attn.k_proj.weight": "model.safetensors",
|
| 532 |
+
"model.layers.29.self_attn.o_proj.biases": "model.safetensors",
|
| 533 |
+
"model.layers.29.self_attn.o_proj.scales": "model.safetensors",
|
| 534 |
+
"model.layers.29.self_attn.o_proj.weight": "model.safetensors",
|
| 535 |
+
"model.layers.29.self_attn.q_proj.biases": "model.safetensors",
|
| 536 |
+
"model.layers.29.self_attn.q_proj.scales": "model.safetensors",
|
| 537 |
+
"model.layers.29.self_attn.q_proj.weight": "model.safetensors",
|
| 538 |
+
"model.layers.29.self_attn.v_proj.biases": "model.safetensors",
|
| 539 |
+
"model.layers.29.self_attn.v_proj.scales": "model.safetensors",
|
| 540 |
+
"model.layers.29.self_attn.v_proj.weight": "model.safetensors",
|
| 541 |
+
"model.layers.3.input_layernorm.weight": "model.safetensors",
|
| 542 |
+
"model.layers.3.mlp.down_proj.biases": "model.safetensors",
|
| 543 |
+
"model.layers.3.mlp.down_proj.scales": "model.safetensors",
|
| 544 |
+
"model.layers.3.mlp.down_proj.weight": "model.safetensors",
|
| 545 |
+
"model.layers.3.mlp.gate_proj.biases": "model.safetensors",
|
| 546 |
+
"model.layers.3.mlp.gate_proj.scales": "model.safetensors",
|
| 547 |
+
"model.layers.3.mlp.gate_proj.weight": "model.safetensors",
|
| 548 |
+
"model.layers.3.mlp.up_proj.biases": "model.safetensors",
|
| 549 |
+
"model.layers.3.mlp.up_proj.scales": "model.safetensors",
|
| 550 |
+
"model.layers.3.mlp.up_proj.weight": "model.safetensors",
|
| 551 |
+
"model.layers.3.post_attention_layernorm.weight": "model.safetensors",
|
| 552 |
+
"model.layers.3.self_attn.k_proj.biases": "model.safetensors",
|
| 553 |
+
"model.layers.3.self_attn.k_proj.scales": "model.safetensors",
|
| 554 |
+
"model.layers.3.self_attn.k_proj.weight": "model.safetensors",
|
| 555 |
+
"model.layers.3.self_attn.o_proj.biases": "model.safetensors",
|
| 556 |
+
"model.layers.3.self_attn.o_proj.scales": "model.safetensors",
|
| 557 |
+
"model.layers.3.self_attn.o_proj.weight": "model.safetensors",
|
| 558 |
+
"model.layers.3.self_attn.q_proj.biases": "model.safetensors",
|
| 559 |
+
"model.layers.3.self_attn.q_proj.scales": "model.safetensors",
|
| 560 |
+
"model.layers.3.self_attn.q_proj.weight": "model.safetensors",
|
| 561 |
+
"model.layers.3.self_attn.v_proj.biases": "model.safetensors",
|
| 562 |
+
"model.layers.3.self_attn.v_proj.scales": "model.safetensors",
|
| 563 |
+
"model.layers.3.self_attn.v_proj.weight": "model.safetensors",
|
| 564 |
+
"model.layers.30.input_layernorm.weight": "model.safetensors",
|
| 565 |
+
"model.layers.30.mlp.down_proj.biases": "model.safetensors",
|
| 566 |
+
"model.layers.30.mlp.down_proj.scales": "model.safetensors",
|
| 567 |
+
"model.layers.30.mlp.down_proj.weight": "model.safetensors",
|
| 568 |
+
"model.layers.30.mlp.gate_proj.biases": "model.safetensors",
|
| 569 |
+
"model.layers.30.mlp.gate_proj.scales": "model.safetensors",
|
| 570 |
+
"model.layers.30.mlp.gate_proj.weight": "model.safetensors",
|
| 571 |
+
"model.layers.30.mlp.up_proj.biases": "model.safetensors",
|
| 572 |
+
"model.layers.30.mlp.up_proj.scales": "model.safetensors",
|
| 573 |
+
"model.layers.30.mlp.up_proj.weight": "model.safetensors",
|
| 574 |
+
"model.layers.30.post_attention_layernorm.weight": "model.safetensors",
|
| 575 |
+
"model.layers.30.self_attn.k_proj.biases": "model.safetensors",
|
| 576 |
+
"model.layers.30.self_attn.k_proj.scales": "model.safetensors",
|
| 577 |
+
"model.layers.30.self_attn.k_proj.weight": "model.safetensors",
|
| 578 |
+
"model.layers.30.self_attn.o_proj.biases": "model.safetensors",
|
| 579 |
+
"model.layers.30.self_attn.o_proj.scales": "model.safetensors",
|
| 580 |
+
"model.layers.30.self_attn.o_proj.weight": "model.safetensors",
|
| 581 |
+
"model.layers.30.self_attn.q_proj.biases": "model.safetensors",
|
| 582 |
+
"model.layers.30.self_attn.q_proj.scales": "model.safetensors",
|
| 583 |
+
"model.layers.30.self_attn.q_proj.weight": "model.safetensors",
|
| 584 |
+
"model.layers.30.self_attn.v_proj.biases": "model.safetensors",
|
| 585 |
+
"model.layers.30.self_attn.v_proj.scales": "model.safetensors",
|
| 586 |
+
"model.layers.30.self_attn.v_proj.weight": "model.safetensors",
|
| 587 |
+
"model.layers.31.input_layernorm.weight": "model.safetensors",
|
| 588 |
+
"model.layers.31.mlp.down_proj.biases": "model.safetensors",
|
| 589 |
+
"model.layers.31.mlp.down_proj.scales": "model.safetensors",
|
| 590 |
+
"model.layers.31.mlp.down_proj.weight": "model.safetensors",
|
| 591 |
+
"model.layers.31.mlp.gate_proj.biases": "model.safetensors",
|
| 592 |
+
"model.layers.31.mlp.gate_proj.scales": "model.safetensors",
|
| 593 |
+
"model.layers.31.mlp.gate_proj.weight": "model.safetensors",
|
| 594 |
+
"model.layers.31.mlp.up_proj.biases": "model.safetensors",
|
| 595 |
+
"model.layers.31.mlp.up_proj.scales": "model.safetensors",
|
| 596 |
+
"model.layers.31.mlp.up_proj.weight": "model.safetensors",
|
| 597 |
+
"model.layers.31.post_attention_layernorm.weight": "model.safetensors",
|
| 598 |
+
"model.layers.31.self_attn.k_proj.biases": "model.safetensors",
|
| 599 |
+
"model.layers.31.self_attn.k_proj.scales": "model.safetensors",
|
| 600 |
+
"model.layers.31.self_attn.k_proj.weight": "model.safetensors",
|
| 601 |
+
"model.layers.31.self_attn.o_proj.biases": "model.safetensors",
|
| 602 |
+
"model.layers.31.self_attn.o_proj.scales": "model.safetensors",
|
| 603 |
+
"model.layers.31.self_attn.o_proj.weight": "model.safetensors",
|
| 604 |
+
"model.layers.31.self_attn.q_proj.biases": "model.safetensors",
|
| 605 |
+
"model.layers.31.self_attn.q_proj.scales": "model.safetensors",
|
| 606 |
+
"model.layers.31.self_attn.q_proj.weight": "model.safetensors",
|
| 607 |
+
"model.layers.31.self_attn.v_proj.biases": "model.safetensors",
|
| 608 |
+
"model.layers.31.self_attn.v_proj.scales": "model.safetensors",
|
| 609 |
+
"model.layers.31.self_attn.v_proj.weight": "model.safetensors",
|
| 610 |
+
"model.layers.4.input_layernorm.weight": "model.safetensors",
|
| 611 |
+
"model.layers.4.mlp.down_proj.biases": "model.safetensors",
|
| 612 |
+
"model.layers.4.mlp.down_proj.scales": "model.safetensors",
|
| 613 |
+
"model.layers.4.mlp.down_proj.weight": "model.safetensors",
|
| 614 |
+
"model.layers.4.mlp.gate_proj.biases": "model.safetensors",
|
| 615 |
+
"model.layers.4.mlp.gate_proj.scales": "model.safetensors",
|
| 616 |
+
"model.layers.4.mlp.gate_proj.weight": "model.safetensors",
|
| 617 |
+
"model.layers.4.mlp.up_proj.biases": "model.safetensors",
|
| 618 |
+
"model.layers.4.mlp.up_proj.scales": "model.safetensors",
|
| 619 |
+
"model.layers.4.mlp.up_proj.weight": "model.safetensors",
|
| 620 |
+
"model.layers.4.post_attention_layernorm.weight": "model.safetensors",
|
| 621 |
+
"model.layers.4.self_attn.k_proj.biases": "model.safetensors",
|
| 622 |
+
"model.layers.4.self_attn.k_proj.scales": "model.safetensors",
|
| 623 |
+
"model.layers.4.self_attn.k_proj.weight": "model.safetensors",
|
| 624 |
+
"model.layers.4.self_attn.o_proj.biases": "model.safetensors",
|
| 625 |
+
"model.layers.4.self_attn.o_proj.scales": "model.safetensors",
|
| 626 |
+
"model.layers.4.self_attn.o_proj.weight": "model.safetensors",
|
| 627 |
+
"model.layers.4.self_attn.q_proj.biases": "model.safetensors",
|
| 628 |
+
"model.layers.4.self_attn.q_proj.scales": "model.safetensors",
|
| 629 |
+
"model.layers.4.self_attn.q_proj.weight": "model.safetensors",
|
| 630 |
+
"model.layers.4.self_attn.v_proj.biases": "model.safetensors",
|
| 631 |
+
"model.layers.4.self_attn.v_proj.scales": "model.safetensors",
|
| 632 |
+
"model.layers.4.self_attn.v_proj.weight": "model.safetensors",
|
| 633 |
+
"model.layers.5.input_layernorm.weight": "model.safetensors",
|
| 634 |
+
"model.layers.5.mlp.down_proj.biases": "model.safetensors",
|
| 635 |
+
"model.layers.5.mlp.down_proj.scales": "model.safetensors",
|
| 636 |
+
"model.layers.5.mlp.down_proj.weight": "model.safetensors",
|
| 637 |
+
"model.layers.5.mlp.gate_proj.biases": "model.safetensors",
|
| 638 |
+
"model.layers.5.mlp.gate_proj.scales": "model.safetensors",
|
| 639 |
+
"model.layers.5.mlp.gate_proj.weight": "model.safetensors",
|
| 640 |
+
"model.layers.5.mlp.up_proj.biases": "model.safetensors",
|
| 641 |
+
"model.layers.5.mlp.up_proj.scales": "model.safetensors",
|
| 642 |
+
"model.layers.5.mlp.up_proj.weight": "model.safetensors",
|
| 643 |
+
"model.layers.5.post_attention_layernorm.weight": "model.safetensors",
|
| 644 |
+
"model.layers.5.self_attn.k_proj.biases": "model.safetensors",
|
| 645 |
+
"model.layers.5.self_attn.k_proj.scales": "model.safetensors",
|
| 646 |
+
"model.layers.5.self_attn.k_proj.weight": "model.safetensors",
|
| 647 |
+
"model.layers.5.self_attn.o_proj.biases": "model.safetensors",
|
| 648 |
+
"model.layers.5.self_attn.o_proj.scales": "model.safetensors",
|
| 649 |
+
"model.layers.5.self_attn.o_proj.weight": "model.safetensors",
|
| 650 |
+
"model.layers.5.self_attn.q_proj.biases": "model.safetensors",
|
| 651 |
+
"model.layers.5.self_attn.q_proj.scales": "model.safetensors",
|
| 652 |
+
"model.layers.5.self_attn.q_proj.weight": "model.safetensors",
|
| 653 |
+
"model.layers.5.self_attn.v_proj.biases": "model.safetensors",
|
| 654 |
+
"model.layers.5.self_attn.v_proj.scales": "model.safetensors",
|
| 655 |
+
"model.layers.5.self_attn.v_proj.weight": "model.safetensors",
|
| 656 |
+
"model.layers.6.input_layernorm.weight": "model.safetensors",
|
| 657 |
+
"model.layers.6.mlp.down_proj.biases": "model.safetensors",
|
| 658 |
+
"model.layers.6.mlp.down_proj.scales": "model.safetensors",
|
| 659 |
+
"model.layers.6.mlp.down_proj.weight": "model.safetensors",
|
| 660 |
+
"model.layers.6.mlp.gate_proj.biases": "model.safetensors",
|
| 661 |
+
"model.layers.6.mlp.gate_proj.scales": "model.safetensors",
|
| 662 |
+
"model.layers.6.mlp.gate_proj.weight": "model.safetensors",
|
| 663 |
+
"model.layers.6.mlp.up_proj.biases": "model.safetensors",
|
| 664 |
+
"model.layers.6.mlp.up_proj.scales": "model.safetensors",
|
| 665 |
+
"model.layers.6.mlp.up_proj.weight": "model.safetensors",
|
| 666 |
+
"model.layers.6.post_attention_layernorm.weight": "model.safetensors",
|
| 667 |
+
"model.layers.6.self_attn.k_proj.biases": "model.safetensors",
|
| 668 |
+
"model.layers.6.self_attn.k_proj.scales": "model.safetensors",
|
| 669 |
+
"model.layers.6.self_attn.k_proj.weight": "model.safetensors",
|
| 670 |
+
"model.layers.6.self_attn.o_proj.biases": "model.safetensors",
|
| 671 |
+
"model.layers.6.self_attn.o_proj.scales": "model.safetensors",
|
| 672 |
+
"model.layers.6.self_attn.o_proj.weight": "model.safetensors",
|
| 673 |
+
"model.layers.6.self_attn.q_proj.biases": "model.safetensors",
|
| 674 |
+
"model.layers.6.self_attn.q_proj.scales": "model.safetensors",
|
| 675 |
+
"model.layers.6.self_attn.q_proj.weight": "model.safetensors",
|
| 676 |
+
"model.layers.6.self_attn.v_proj.biases": "model.safetensors",
|
| 677 |
+
"model.layers.6.self_attn.v_proj.scales": "model.safetensors",
|
| 678 |
+
"model.layers.6.self_attn.v_proj.weight": "model.safetensors",
|
| 679 |
+
"model.layers.7.input_layernorm.weight": "model.safetensors",
|
| 680 |
+
"model.layers.7.mlp.down_proj.biases": "model.safetensors",
|
| 681 |
+
"model.layers.7.mlp.down_proj.scales": "model.safetensors",
|
| 682 |
+
"model.layers.7.mlp.down_proj.weight": "model.safetensors",
|
| 683 |
+
"model.layers.7.mlp.gate_proj.biases": "model.safetensors",
|
| 684 |
+
"model.layers.7.mlp.gate_proj.scales": "model.safetensors",
|
| 685 |
+
"model.layers.7.mlp.gate_proj.weight": "model.safetensors",
|
| 686 |
+
"model.layers.7.mlp.up_proj.biases": "model.safetensors",
|
| 687 |
+
"model.layers.7.mlp.up_proj.scales": "model.safetensors",
|
| 688 |
+
"model.layers.7.mlp.up_proj.weight": "model.safetensors",
|
| 689 |
+
"model.layers.7.post_attention_layernorm.weight": "model.safetensors",
|
| 690 |
+
"model.layers.7.self_attn.k_proj.biases": "model.safetensors",
|
| 691 |
+
"model.layers.7.self_attn.k_proj.scales": "model.safetensors",
|
| 692 |
+
"model.layers.7.self_attn.k_proj.weight": "model.safetensors",
|
| 693 |
+
"model.layers.7.self_attn.o_proj.biases": "model.safetensors",
|
| 694 |
+
"model.layers.7.self_attn.o_proj.scales": "model.safetensors",
|
| 695 |
+
"model.layers.7.self_attn.o_proj.weight": "model.safetensors",
|
| 696 |
+
"model.layers.7.self_attn.q_proj.biases": "model.safetensors",
|
| 697 |
+
"model.layers.7.self_attn.q_proj.scales": "model.safetensors",
|
| 698 |
+
"model.layers.7.self_attn.q_proj.weight": "model.safetensors",
|
| 699 |
+
"model.layers.7.self_attn.v_proj.biases": "model.safetensors",
|
| 700 |
+
"model.layers.7.self_attn.v_proj.scales": "model.safetensors",
|
| 701 |
+
"model.layers.7.self_attn.v_proj.weight": "model.safetensors",
|
| 702 |
+
"model.layers.8.input_layernorm.weight": "model.safetensors",
|
| 703 |
+
"model.layers.8.mlp.down_proj.biases": "model.safetensors",
|
| 704 |
+
"model.layers.8.mlp.down_proj.scales": "model.safetensors",
|
| 705 |
+
"model.layers.8.mlp.down_proj.weight": "model.safetensors",
|
| 706 |
+
"model.layers.8.mlp.gate_proj.biases": "model.safetensors",
|
| 707 |
+
"model.layers.8.mlp.gate_proj.scales": "model.safetensors",
|
| 708 |
+
"model.layers.8.mlp.gate_proj.weight": "model.safetensors",
|
| 709 |
+
"model.layers.8.mlp.up_proj.biases": "model.safetensors",
|
| 710 |
+
"model.layers.8.mlp.up_proj.scales": "model.safetensors",
|
| 711 |
+
"model.layers.8.mlp.up_proj.weight": "model.safetensors",
|
| 712 |
+
"model.layers.8.post_attention_layernorm.weight": "model.safetensors",
|
| 713 |
+
"model.layers.8.self_attn.k_proj.biases": "model.safetensors",
|
| 714 |
+
"model.layers.8.self_attn.k_proj.scales": "model.safetensors",
|
| 715 |
+
"model.layers.8.self_attn.k_proj.weight": "model.safetensors",
|
| 716 |
+
"model.layers.8.self_attn.o_proj.biases": "model.safetensors",
|
| 717 |
+
"model.layers.8.self_attn.o_proj.scales": "model.safetensors",
|
| 718 |
+
"model.layers.8.self_attn.o_proj.weight": "model.safetensors",
|
| 719 |
+
"model.layers.8.self_attn.q_proj.biases": "model.safetensors",
|
| 720 |
+
"model.layers.8.self_attn.q_proj.scales": "model.safetensors",
|
| 721 |
+
"model.layers.8.self_attn.q_proj.weight": "model.safetensors",
|
| 722 |
+
"model.layers.8.self_attn.v_proj.biases": "model.safetensors",
|
| 723 |
+
"model.layers.8.self_attn.v_proj.scales": "model.safetensors",
|
| 724 |
+
"model.layers.8.self_attn.v_proj.weight": "model.safetensors",
|
| 725 |
+
"model.layers.9.input_layernorm.weight": "model.safetensors",
|
| 726 |
+
"model.layers.9.mlp.down_proj.biases": "model.safetensors",
|
| 727 |
+
"model.layers.9.mlp.down_proj.scales": "model.safetensors",
|
| 728 |
+
"model.layers.9.mlp.down_proj.weight": "model.safetensors",
|
| 729 |
+
"model.layers.9.mlp.gate_proj.biases": "model.safetensors",
|
| 730 |
+
"model.layers.9.mlp.gate_proj.scales": "model.safetensors",
|
| 731 |
+
"model.layers.9.mlp.gate_proj.weight": "model.safetensors",
|
| 732 |
+
"model.layers.9.mlp.up_proj.biases": "model.safetensors",
|
| 733 |
+
"model.layers.9.mlp.up_proj.scales": "model.safetensors",
|
| 734 |
+
"model.layers.9.mlp.up_proj.weight": "model.safetensors",
|
| 735 |
+
"model.layers.9.post_attention_layernorm.weight": "model.safetensors",
|
| 736 |
+
"model.layers.9.self_attn.k_proj.biases": "model.safetensors",
|
| 737 |
+
"model.layers.9.self_attn.k_proj.scales": "model.safetensors",
|
| 738 |
+
"model.layers.9.self_attn.k_proj.weight": "model.safetensors",
|
| 739 |
+
"model.layers.9.self_attn.o_proj.biases": "model.safetensors",
|
| 740 |
+
"model.layers.9.self_attn.o_proj.scales": "model.safetensors",
|
| 741 |
+
"model.layers.9.self_attn.o_proj.weight": "model.safetensors",
|
| 742 |
+
"model.layers.9.self_attn.q_proj.biases": "model.safetensors",
|
| 743 |
+
"model.layers.9.self_attn.q_proj.scales": "model.safetensors",
|
| 744 |
+
"model.layers.9.self_attn.q_proj.weight": "model.safetensors",
|
| 745 |
+
"model.layers.9.self_attn.v_proj.biases": "model.safetensors",
|
| 746 |
+
"model.layers.9.self_attn.v_proj.scales": "model.safetensors",
|
| 747 |
+
"model.layers.9.self_attn.v_proj.weight": "model.safetensors",
|
| 748 |
+
"model.norm.weight": "model.safetensors"
|
| 749 |
+
}
|
| 750 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<[begin▁of▁sentence]>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<[end▁of▁sentence]>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<[PAD▁TOKEN]>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "<[SEP▁TOKEN]>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:db6520146c388c495a98bbea62ff6d00c0a8935bed33622e33bb33ec71aaafed
|
| 3 |
+
size 11891696
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,1037 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<[begin▁of▁sentence]>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<[PAD▁TOKEN]>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "<[end▁of▁sentence]>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<[UNK_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"4": {
|
| 36 |
+
"content": "<[CLS_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"5": {
|
| 44 |
+
"content": "<[MASK_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"6": {
|
| 52 |
+
"content": "<[SEP▁TOKEN]>",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"7": {
|
| 60 |
+
"content": "<[PLHD7_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"8": {
|
| 68 |
+
"content": "<[PLHD8_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 69 |
+
"lstrip": false,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": false,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"9": {
|
| 76 |
+
"content": "<[PLHD9_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 77 |
+
"lstrip": false,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": false,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"10": {
|
| 84 |
+
"content": "<[PLHD10_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 85 |
+
"lstrip": false,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": false,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"11": {
|
| 92 |
+
"content": "<[PLHD11_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 93 |
+
"lstrip": false,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": false,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"12": {
|
| 100 |
+
"content": "<[PLHD12_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 101 |
+
"lstrip": false,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": false,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"13": {
|
| 108 |
+
"content": "<[PLHD13_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 109 |
+
"lstrip": false,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": false,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"14": {
|
| 116 |
+
"content": "<[PLHD14_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 117 |
+
"lstrip": false,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": false,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"15": {
|
| 124 |
+
"content": "<[PLHD15_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 125 |
+
"lstrip": false,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": false,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"16": {
|
| 132 |
+
"content": "<[PLHD16_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 133 |
+
"lstrip": false,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": false,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"17": {
|
| 140 |
+
"content": "<[PLHD17_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 141 |
+
"lstrip": false,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": false,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"18": {
|
| 148 |
+
"content": "<[PLHD18_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 149 |
+
"lstrip": false,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": false,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"19": {
|
| 156 |
+
"content": "<[PLHD19_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 157 |
+
"lstrip": false,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": false,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"20": {
|
| 164 |
+
"content": "<[PLHD20_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 165 |
+
"lstrip": false,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": false,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"21": {
|
| 172 |
+
"content": "<[PLHD21_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 173 |
+
"lstrip": false,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": false,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
},
|
| 179 |
+
"22": {
|
| 180 |
+
"content": "<[PLHD22_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 181 |
+
"lstrip": false,
|
| 182 |
+
"normalized": false,
|
| 183 |
+
"rstrip": false,
|
| 184 |
+
"single_word": false,
|
| 185 |
+
"special": true
|
| 186 |
+
},
|
| 187 |
+
"23": {
|
| 188 |
+
"content": "<[PLHD23_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 189 |
+
"lstrip": false,
|
| 190 |
+
"normalized": false,
|
| 191 |
+
"rstrip": false,
|
| 192 |
+
"single_word": false,
|
| 193 |
+
"special": true
|
| 194 |
+
},
|
| 195 |
+
"24": {
|
| 196 |
+
"content": "<[PLHD24_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 197 |
+
"lstrip": false,
|
| 198 |
+
"normalized": false,
|
| 199 |
+
"rstrip": false,
|
| 200 |
+
"single_word": false,
|
| 201 |
+
"special": true
|
| 202 |
+
},
|
| 203 |
+
"25": {
|
| 204 |
+
"content": "<[PLHD25_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 205 |
+
"lstrip": false,
|
| 206 |
+
"normalized": false,
|
| 207 |
+
"rstrip": false,
|
| 208 |
+
"single_word": false,
|
| 209 |
+
"special": true
|
| 210 |
+
},
|
| 211 |
+
"26": {
|
| 212 |
+
"content": "<[PLHD26_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 213 |
+
"lstrip": false,
|
| 214 |
+
"normalized": false,
|
| 215 |
+
"rstrip": false,
|
| 216 |
+
"single_word": false,
|
| 217 |
+
"special": true
|
| 218 |
+
},
|
| 219 |
+
"27": {
|
| 220 |
+
"content": "<[PLHD27_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 221 |
+
"lstrip": false,
|
| 222 |
+
"normalized": false,
|
| 223 |
+
"rstrip": false,
|
| 224 |
+
"single_word": false,
|
| 225 |
+
"special": true
|
| 226 |
+
},
|
| 227 |
+
"28": {
|
| 228 |
+
"content": "<[PLHD28_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 229 |
+
"lstrip": false,
|
| 230 |
+
"normalized": false,
|
| 231 |
+
"rstrip": false,
|
| 232 |
+
"single_word": false,
|
| 233 |
+
"special": true
|
| 234 |
+
},
|
| 235 |
+
"29": {
|
| 236 |
+
"content": "<[PLHD29_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 237 |
+
"lstrip": false,
|
| 238 |
+
"normalized": false,
|
| 239 |
+
"rstrip": false,
|
| 240 |
+
"single_word": false,
|
| 241 |
+
"special": true
|
| 242 |
+
},
|
| 243 |
+
"30": {
|
| 244 |
+
"content": "<[PLHD30_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 245 |
+
"lstrip": false,
|
| 246 |
+
"normalized": false,
|
| 247 |
+
"rstrip": false,
|
| 248 |
+
"single_word": false,
|
| 249 |
+
"special": true
|
| 250 |
+
},
|
| 251 |
+
"31": {
|
| 252 |
+
"content": "<[PLHD31_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 253 |
+
"lstrip": false,
|
| 254 |
+
"normalized": false,
|
| 255 |
+
"rstrip": false,
|
| 256 |
+
"single_word": false,
|
| 257 |
+
"special": true
|
| 258 |
+
},
|
| 259 |
+
"32": {
|
| 260 |
+
"content": "<[PLHD32_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 261 |
+
"lstrip": false,
|
| 262 |
+
"normalized": false,
|
| 263 |
+
"rstrip": false,
|
| 264 |
+
"single_word": false,
|
| 265 |
+
"special": true
|
| 266 |
+
},
|
| 267 |
+
"33": {
|
| 268 |
+
"content": "<[PLHD33_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 269 |
+
"lstrip": false,
|
| 270 |
+
"normalized": false,
|
| 271 |
+
"rstrip": false,
|
| 272 |
+
"single_word": false,
|
| 273 |
+
"special": true
|
| 274 |
+
},
|
| 275 |
+
"34": {
|
| 276 |
+
"content": "<[PLHD34_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 277 |
+
"lstrip": false,
|
| 278 |
+
"normalized": false,
|
| 279 |
+
"rstrip": false,
|
| 280 |
+
"single_word": false,
|
| 281 |
+
"special": true
|
| 282 |
+
},
|
| 283 |
+
"35": {
|
| 284 |
+
"content": "<[PLHD35_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 285 |
+
"lstrip": false,
|
| 286 |
+
"normalized": false,
|
| 287 |
+
"rstrip": false,
|
| 288 |
+
"single_word": false,
|
| 289 |
+
"special": true
|
| 290 |
+
},
|
| 291 |
+
"36": {
|
| 292 |
+
"content": "<[PLHD36_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 293 |
+
"lstrip": false,
|
| 294 |
+
"normalized": false,
|
| 295 |
+
"rstrip": false,
|
| 296 |
+
"single_word": false,
|
| 297 |
+
"special": true
|
| 298 |
+
},
|
| 299 |
+
"37": {
|
| 300 |
+
"content": "<[PLHD37_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 301 |
+
"lstrip": false,
|
| 302 |
+
"normalized": false,
|
| 303 |
+
"rstrip": false,
|
| 304 |
+
"single_word": false,
|
| 305 |
+
"special": true
|
| 306 |
+
},
|
| 307 |
+
"38": {
|
| 308 |
+
"content": "<[PLHD38_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 309 |
+
"lstrip": false,
|
| 310 |
+
"normalized": false,
|
| 311 |
+
"rstrip": false,
|
| 312 |
+
"single_word": false,
|
| 313 |
+
"special": true
|
| 314 |
+
},
|
| 315 |
+
"39": {
|
| 316 |
+
"content": "<[PLHD39_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 317 |
+
"lstrip": false,
|
| 318 |
+
"normalized": false,
|
| 319 |
+
"rstrip": false,
|
| 320 |
+
"single_word": false,
|
| 321 |
+
"special": true
|
| 322 |
+
},
|
| 323 |
+
"40": {
|
| 324 |
+
"content": "<[PLHD40_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 325 |
+
"lstrip": false,
|
| 326 |
+
"normalized": false,
|
| 327 |
+
"rstrip": false,
|
| 328 |
+
"single_word": false,
|
| 329 |
+
"special": true
|
| 330 |
+
},
|
| 331 |
+
"41": {
|
| 332 |
+
"content": "<[PLHD41_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 333 |
+
"lstrip": false,
|
| 334 |
+
"normalized": false,
|
| 335 |
+
"rstrip": false,
|
| 336 |
+
"single_word": false,
|
| 337 |
+
"special": true
|
| 338 |
+
},
|
| 339 |
+
"42": {
|
| 340 |
+
"content": "<[PLHD42_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 341 |
+
"lstrip": false,
|
| 342 |
+
"normalized": false,
|
| 343 |
+
"rstrip": false,
|
| 344 |
+
"single_word": false,
|
| 345 |
+
"special": true
|
| 346 |
+
},
|
| 347 |
+
"43": {
|
| 348 |
+
"content": "<[PLHD43_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 349 |
+
"lstrip": false,
|
| 350 |
+
"normalized": false,
|
| 351 |
+
"rstrip": false,
|
| 352 |
+
"single_word": false,
|
| 353 |
+
"special": true
|
| 354 |
+
},
|
| 355 |
+
"44": {
|
| 356 |
+
"content": "<[PLHD44_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 357 |
+
"lstrip": false,
|
| 358 |
+
"normalized": false,
|
| 359 |
+
"rstrip": false,
|
| 360 |
+
"single_word": false,
|
| 361 |
+
"special": true
|
| 362 |
+
},
|
| 363 |
+
"45": {
|
| 364 |
+
"content": "<[PLHD45_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 365 |
+
"lstrip": false,
|
| 366 |
+
"normalized": false,
|
| 367 |
+
"rstrip": false,
|
| 368 |
+
"single_word": false,
|
| 369 |
+
"special": true
|
| 370 |
+
},
|
| 371 |
+
"46": {
|
| 372 |
+
"content": "<[PLHD46_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 373 |
+
"lstrip": false,
|
| 374 |
+
"normalized": false,
|
| 375 |
+
"rstrip": false,
|
| 376 |
+
"single_word": false,
|
| 377 |
+
"special": true
|
| 378 |
+
},
|
| 379 |
+
"47": {
|
| 380 |
+
"content": "<[PLHD47_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 381 |
+
"lstrip": false,
|
| 382 |
+
"normalized": false,
|
| 383 |
+
"rstrip": false,
|
| 384 |
+
"single_word": false,
|
| 385 |
+
"special": true
|
| 386 |
+
},
|
| 387 |
+
"48": {
|
| 388 |
+
"content": "<[PLHD48_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 389 |
+
"lstrip": false,
|
| 390 |
+
"normalized": false,
|
| 391 |
+
"rstrip": false,
|
| 392 |
+
"single_word": false,
|
| 393 |
+
"special": true
|
| 394 |
+
},
|
| 395 |
+
"49": {
|
| 396 |
+
"content": "<[PLHD49_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 397 |
+
"lstrip": false,
|
| 398 |
+
"normalized": false,
|
| 399 |
+
"rstrip": false,
|
| 400 |
+
"single_word": false,
|
| 401 |
+
"special": true
|
| 402 |
+
},
|
| 403 |
+
"50": {
|
| 404 |
+
"content": "<[PLHD50_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 405 |
+
"lstrip": false,
|
| 406 |
+
"normalized": false,
|
| 407 |
+
"rstrip": false,
|
| 408 |
+
"single_word": false,
|
| 409 |
+
"special": true
|
| 410 |
+
},
|
| 411 |
+
"51": {
|
| 412 |
+
"content": "<[PLHD51_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 413 |
+
"lstrip": false,
|
| 414 |
+
"normalized": false,
|
| 415 |
+
"rstrip": false,
|
| 416 |
+
"single_word": false,
|
| 417 |
+
"special": true
|
| 418 |
+
},
|
| 419 |
+
"52": {
|
| 420 |
+
"content": "<[PLHD52_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 421 |
+
"lstrip": false,
|
| 422 |
+
"normalized": false,
|
| 423 |
+
"rstrip": false,
|
| 424 |
+
"single_word": false,
|
| 425 |
+
"special": true
|
| 426 |
+
},
|
| 427 |
+
"53": {
|
| 428 |
+
"content": "<[PLHD53_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 429 |
+
"lstrip": false,
|
| 430 |
+
"normalized": false,
|
| 431 |
+
"rstrip": false,
|
| 432 |
+
"single_word": false,
|
| 433 |
+
"special": true
|
| 434 |
+
},
|
| 435 |
+
"54": {
|
| 436 |
+
"content": "<[PLHD54_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 437 |
+
"lstrip": false,
|
| 438 |
+
"normalized": false,
|
| 439 |
+
"rstrip": false,
|
| 440 |
+
"single_word": false,
|
| 441 |
+
"special": true
|
| 442 |
+
},
|
| 443 |
+
"55": {
|
| 444 |
+
"content": "<[PLHD55_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 445 |
+
"lstrip": false,
|
| 446 |
+
"normalized": false,
|
| 447 |
+
"rstrip": false,
|
| 448 |
+
"single_word": false,
|
| 449 |
+
"special": true
|
| 450 |
+
},
|
| 451 |
+
"56": {
|
| 452 |
+
"content": "<[PLHD56_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 453 |
+
"lstrip": false,
|
| 454 |
+
"normalized": false,
|
| 455 |
+
"rstrip": false,
|
| 456 |
+
"single_word": false,
|
| 457 |
+
"special": true
|
| 458 |
+
},
|
| 459 |
+
"57": {
|
| 460 |
+
"content": "<[PLHD57_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 461 |
+
"lstrip": false,
|
| 462 |
+
"normalized": false,
|
| 463 |
+
"rstrip": false,
|
| 464 |
+
"single_word": false,
|
| 465 |
+
"special": true
|
| 466 |
+
},
|
| 467 |
+
"58": {
|
| 468 |
+
"content": "<[PLHD58_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 469 |
+
"lstrip": false,
|
| 470 |
+
"normalized": false,
|
| 471 |
+
"rstrip": false,
|
| 472 |
+
"single_word": false,
|
| 473 |
+
"special": true
|
| 474 |
+
},
|
| 475 |
+
"59": {
|
| 476 |
+
"content": "<[PLHD59_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 477 |
+
"lstrip": false,
|
| 478 |
+
"normalized": false,
|
| 479 |
+
"rstrip": false,
|
| 480 |
+
"single_word": false,
|
| 481 |
+
"special": true
|
| 482 |
+
},
|
| 483 |
+
"60": {
|
| 484 |
+
"content": "<[PLHD60_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 485 |
+
"lstrip": false,
|
| 486 |
+
"normalized": false,
|
| 487 |
+
"rstrip": false,
|
| 488 |
+
"single_word": false,
|
| 489 |
+
"special": true
|
| 490 |
+
},
|
| 491 |
+
"61": {
|
| 492 |
+
"content": "<[PLHD61_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 493 |
+
"lstrip": false,
|
| 494 |
+
"normalized": false,
|
| 495 |
+
"rstrip": false,
|
| 496 |
+
"single_word": false,
|
| 497 |
+
"special": true
|
| 498 |
+
},
|
| 499 |
+
"62": {
|
| 500 |
+
"content": "<[PLHD62_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 501 |
+
"lstrip": false,
|
| 502 |
+
"normalized": false,
|
| 503 |
+
"rstrip": false,
|
| 504 |
+
"single_word": false,
|
| 505 |
+
"special": true
|
| 506 |
+
},
|
| 507 |
+
"63": {
|
| 508 |
+
"content": "<[PLHD63_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 509 |
+
"lstrip": false,
|
| 510 |
+
"normalized": false,
|
| 511 |
+
"rstrip": false,
|
| 512 |
+
"single_word": false,
|
| 513 |
+
"special": true
|
| 514 |
+
},
|
| 515 |
+
"64": {
|
| 516 |
+
"content": "<[PLHD64_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 517 |
+
"lstrip": false,
|
| 518 |
+
"normalized": false,
|
| 519 |
+
"rstrip": false,
|
| 520 |
+
"single_word": false,
|
| 521 |
+
"special": true
|
| 522 |
+
},
|
| 523 |
+
"65": {
|
| 524 |
+
"content": "<[PLHD65_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 525 |
+
"lstrip": false,
|
| 526 |
+
"normalized": false,
|
| 527 |
+
"rstrip": false,
|
| 528 |
+
"single_word": false,
|
| 529 |
+
"special": true
|
| 530 |
+
},
|
| 531 |
+
"66": {
|
| 532 |
+
"content": "<[PLHD66_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 533 |
+
"lstrip": false,
|
| 534 |
+
"normalized": false,
|
| 535 |
+
"rstrip": false,
|
| 536 |
+
"single_word": false,
|
| 537 |
+
"special": true
|
| 538 |
+
},
|
| 539 |
+
"67": {
|
| 540 |
+
"content": "<[PLHD67_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 541 |
+
"lstrip": false,
|
| 542 |
+
"normalized": false,
|
| 543 |
+
"rstrip": false,
|
| 544 |
+
"single_word": false,
|
| 545 |
+
"special": true
|
| 546 |
+
},
|
| 547 |
+
"68": {
|
| 548 |
+
"content": "<[PLHD68_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 549 |
+
"lstrip": false,
|
| 550 |
+
"normalized": false,
|
| 551 |
+
"rstrip": false,
|
| 552 |
+
"single_word": false,
|
| 553 |
+
"special": true
|
| 554 |
+
},
|
| 555 |
+
"69": {
|
| 556 |
+
"content": "<[PLHD69_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 557 |
+
"lstrip": false,
|
| 558 |
+
"normalized": false,
|
| 559 |
+
"rstrip": false,
|
| 560 |
+
"single_word": false,
|
| 561 |
+
"special": true
|
| 562 |
+
},
|
| 563 |
+
"70": {
|
| 564 |
+
"content": "<[PLHD70_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 565 |
+
"lstrip": false,
|
| 566 |
+
"normalized": false,
|
| 567 |
+
"rstrip": false,
|
| 568 |
+
"single_word": false,
|
| 569 |
+
"special": true
|
| 570 |
+
},
|
| 571 |
+
"71": {
|
| 572 |
+
"content": "<[PLHD71_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 573 |
+
"lstrip": false,
|
| 574 |
+
"normalized": false,
|
| 575 |
+
"rstrip": false,
|
| 576 |
+
"single_word": false,
|
| 577 |
+
"special": true
|
| 578 |
+
},
|
| 579 |
+
"72": {
|
| 580 |
+
"content": "<[PLHD72_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 581 |
+
"lstrip": false,
|
| 582 |
+
"normalized": false,
|
| 583 |
+
"rstrip": false,
|
| 584 |
+
"single_word": false,
|
| 585 |
+
"special": true
|
| 586 |
+
},
|
| 587 |
+
"73": {
|
| 588 |
+
"content": "<[PLHD73_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 589 |
+
"lstrip": false,
|
| 590 |
+
"normalized": false,
|
| 591 |
+
"rstrip": false,
|
| 592 |
+
"single_word": false,
|
| 593 |
+
"special": true
|
| 594 |
+
},
|
| 595 |
+
"74": {
|
| 596 |
+
"content": "<[PLHD74_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 597 |
+
"lstrip": false,
|
| 598 |
+
"normalized": false,
|
| 599 |
+
"rstrip": false,
|
| 600 |
+
"single_word": false,
|
| 601 |
+
"special": true
|
| 602 |
+
},
|
| 603 |
+
"75": {
|
| 604 |
+
"content": "<[PLHD75_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 605 |
+
"lstrip": false,
|
| 606 |
+
"normalized": false,
|
| 607 |
+
"rstrip": false,
|
| 608 |
+
"single_word": false,
|
| 609 |
+
"special": true
|
| 610 |
+
},
|
| 611 |
+
"76": {
|
| 612 |
+
"content": "<[PLHD76_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 613 |
+
"lstrip": false,
|
| 614 |
+
"normalized": false,
|
| 615 |
+
"rstrip": false,
|
| 616 |
+
"single_word": false,
|
| 617 |
+
"special": true
|
| 618 |
+
},
|
| 619 |
+
"77": {
|
| 620 |
+
"content": "<[PLHD77_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 621 |
+
"lstrip": false,
|
| 622 |
+
"normalized": false,
|
| 623 |
+
"rstrip": false,
|
| 624 |
+
"single_word": false,
|
| 625 |
+
"special": true
|
| 626 |
+
},
|
| 627 |
+
"78": {
|
| 628 |
+
"content": "<[PLHD78_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 629 |
+
"lstrip": false,
|
| 630 |
+
"normalized": false,
|
| 631 |
+
"rstrip": false,
|
| 632 |
+
"single_word": false,
|
| 633 |
+
"special": true
|
| 634 |
+
},
|
| 635 |
+
"79": {
|
| 636 |
+
"content": "<[PLHD79_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 637 |
+
"lstrip": false,
|
| 638 |
+
"normalized": false,
|
| 639 |
+
"rstrip": false,
|
| 640 |
+
"single_word": false,
|
| 641 |
+
"special": true
|
| 642 |
+
},
|
| 643 |
+
"80": {
|
| 644 |
+
"content": "<[PLHD80_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 645 |
+
"lstrip": false,
|
| 646 |
+
"normalized": false,
|
| 647 |
+
"rstrip": false,
|
| 648 |
+
"single_word": false,
|
| 649 |
+
"special": true
|
| 650 |
+
},
|
| 651 |
+
"81": {
|
| 652 |
+
"content": "<[PLHD81_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 653 |
+
"lstrip": false,
|
| 654 |
+
"normalized": false,
|
| 655 |
+
"rstrip": false,
|
| 656 |
+
"single_word": false,
|
| 657 |
+
"special": true
|
| 658 |
+
},
|
| 659 |
+
"82": {
|
| 660 |
+
"content": "<[PLHD82_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 661 |
+
"lstrip": false,
|
| 662 |
+
"normalized": false,
|
| 663 |
+
"rstrip": false,
|
| 664 |
+
"single_word": false,
|
| 665 |
+
"special": true
|
| 666 |
+
},
|
| 667 |
+
"83": {
|
| 668 |
+
"content": "<[PLHD83_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 669 |
+
"lstrip": false,
|
| 670 |
+
"normalized": false,
|
| 671 |
+
"rstrip": false,
|
| 672 |
+
"single_word": false,
|
| 673 |
+
"special": true
|
| 674 |
+
},
|
| 675 |
+
"84": {
|
| 676 |
+
"content": "<[PLHD84_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 677 |
+
"lstrip": false,
|
| 678 |
+
"normalized": false,
|
| 679 |
+
"rstrip": false,
|
| 680 |
+
"single_word": false,
|
| 681 |
+
"special": true
|
| 682 |
+
},
|
| 683 |
+
"85": {
|
| 684 |
+
"content": "<[PLHD85_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 685 |
+
"lstrip": false,
|
| 686 |
+
"normalized": false,
|
| 687 |
+
"rstrip": false,
|
| 688 |
+
"single_word": false,
|
| 689 |
+
"special": true
|
| 690 |
+
},
|
| 691 |
+
"86": {
|
| 692 |
+
"content": "<[PLHD86_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 693 |
+
"lstrip": false,
|
| 694 |
+
"normalized": false,
|
| 695 |
+
"rstrip": false,
|
| 696 |
+
"single_word": false,
|
| 697 |
+
"special": true
|
| 698 |
+
},
|
| 699 |
+
"87": {
|
| 700 |
+
"content": "<[PLHD87_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 701 |
+
"lstrip": false,
|
| 702 |
+
"normalized": false,
|
| 703 |
+
"rstrip": false,
|
| 704 |
+
"single_word": false,
|
| 705 |
+
"special": true
|
| 706 |
+
},
|
| 707 |
+
"88": {
|
| 708 |
+
"content": "<[PLHD88_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 709 |
+
"lstrip": false,
|
| 710 |
+
"normalized": false,
|
| 711 |
+
"rstrip": false,
|
| 712 |
+
"single_word": false,
|
| 713 |
+
"special": true
|
| 714 |
+
},
|
| 715 |
+
"89": {
|
| 716 |
+
"content": "<[PLHD89_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 717 |
+
"lstrip": false,
|
| 718 |
+
"normalized": false,
|
| 719 |
+
"rstrip": false,
|
| 720 |
+
"single_word": false,
|
| 721 |
+
"special": true
|
| 722 |
+
},
|
| 723 |
+
"90": {
|
| 724 |
+
"content": "<[PLHD90_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 725 |
+
"lstrip": false,
|
| 726 |
+
"normalized": false,
|
| 727 |
+
"rstrip": false,
|
| 728 |
+
"single_word": false,
|
| 729 |
+
"special": true
|
| 730 |
+
},
|
| 731 |
+
"91": {
|
| 732 |
+
"content": "<[PLHD91_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 733 |
+
"lstrip": false,
|
| 734 |
+
"normalized": false,
|
| 735 |
+
"rstrip": false,
|
| 736 |
+
"single_word": false,
|
| 737 |
+
"special": true
|
| 738 |
+
},
|
| 739 |
+
"92": {
|
| 740 |
+
"content": "<[PLHD92_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 741 |
+
"lstrip": false,
|
| 742 |
+
"normalized": false,
|
| 743 |
+
"rstrip": false,
|
| 744 |
+
"single_word": false,
|
| 745 |
+
"special": true
|
| 746 |
+
},
|
| 747 |
+
"93": {
|
| 748 |
+
"content": "<[PLHD93_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 749 |
+
"lstrip": false,
|
| 750 |
+
"normalized": false,
|
| 751 |
+
"rstrip": false,
|
| 752 |
+
"single_word": false,
|
| 753 |
+
"special": true
|
| 754 |
+
},
|
| 755 |
+
"94": {
|
| 756 |
+
"content": "<[PLHD94_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 757 |
+
"lstrip": false,
|
| 758 |
+
"normalized": false,
|
| 759 |
+
"rstrip": false,
|
| 760 |
+
"single_word": false,
|
| 761 |
+
"special": true
|
| 762 |
+
},
|
| 763 |
+
"95": {
|
| 764 |
+
"content": "<[PLHD95_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 765 |
+
"lstrip": false,
|
| 766 |
+
"normalized": false,
|
| 767 |
+
"rstrip": false,
|
| 768 |
+
"single_word": false,
|
| 769 |
+
"special": true
|
| 770 |
+
},
|
| 771 |
+
"96": {
|
| 772 |
+
"content": "<[PLHD96_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 773 |
+
"lstrip": false,
|
| 774 |
+
"normalized": false,
|
| 775 |
+
"rstrip": false,
|
| 776 |
+
"single_word": false,
|
| 777 |
+
"special": true
|
| 778 |
+
},
|
| 779 |
+
"97": {
|
| 780 |
+
"content": "<[PLHD97_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 781 |
+
"lstrip": false,
|
| 782 |
+
"normalized": false,
|
| 783 |
+
"rstrip": false,
|
| 784 |
+
"single_word": false,
|
| 785 |
+
"special": true
|
| 786 |
+
},
|
| 787 |
+
"98": {
|
| 788 |
+
"content": "<[PLHD98_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 789 |
+
"lstrip": false,
|
| 790 |
+
"normalized": false,
|
| 791 |
+
"rstrip": false,
|
| 792 |
+
"single_word": false,
|
| 793 |
+
"special": true
|
| 794 |
+
},
|
| 795 |
+
"99": {
|
| 796 |
+
"content": "<[PLHD99_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 797 |
+
"lstrip": false,
|
| 798 |
+
"normalized": false,
|
| 799 |
+
"rstrip": false,
|
| 800 |
+
"single_word": false,
|
| 801 |
+
"special": true
|
| 802 |
+
},
|
| 803 |
+
"100": {
|
| 804 |
+
"content": "<[PLHD100_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 805 |
+
"lstrip": false,
|
| 806 |
+
"normalized": false,
|
| 807 |
+
"rstrip": false,
|
| 808 |
+
"single_word": false,
|
| 809 |
+
"special": true
|
| 810 |
+
},
|
| 811 |
+
"101": {
|
| 812 |
+
"content": "<[PLHD101_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 813 |
+
"lstrip": false,
|
| 814 |
+
"normalized": false,
|
| 815 |
+
"rstrip": false,
|
| 816 |
+
"single_word": false,
|
| 817 |
+
"special": true
|
| 818 |
+
},
|
| 819 |
+
"102": {
|
| 820 |
+
"content": "<[PLHD102_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 821 |
+
"lstrip": false,
|
| 822 |
+
"normalized": false,
|
| 823 |
+
"rstrip": false,
|
| 824 |
+
"single_word": false,
|
| 825 |
+
"special": true
|
| 826 |
+
},
|
| 827 |
+
"103": {
|
| 828 |
+
"content": "<[PLHD103_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 829 |
+
"lstrip": false,
|
| 830 |
+
"normalized": false,
|
| 831 |
+
"rstrip": false,
|
| 832 |
+
"single_word": false,
|
| 833 |
+
"special": true
|
| 834 |
+
},
|
| 835 |
+
"104": {
|
| 836 |
+
"content": "<[PLHD104_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 837 |
+
"lstrip": false,
|
| 838 |
+
"normalized": false,
|
| 839 |
+
"rstrip": false,
|
| 840 |
+
"single_word": false,
|
| 841 |
+
"special": true
|
| 842 |
+
},
|
| 843 |
+
"105": {
|
| 844 |
+
"content": "<[PLHD105_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 845 |
+
"lstrip": false,
|
| 846 |
+
"normalized": false,
|
| 847 |
+
"rstrip": false,
|
| 848 |
+
"single_word": false,
|
| 849 |
+
"special": true
|
| 850 |
+
},
|
| 851 |
+
"106": {
|
| 852 |
+
"content": "<[PLHD106_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 853 |
+
"lstrip": false,
|
| 854 |
+
"normalized": false,
|
| 855 |
+
"rstrip": false,
|
| 856 |
+
"single_word": false,
|
| 857 |
+
"special": true
|
| 858 |
+
},
|
| 859 |
+
"107": {
|
| 860 |
+
"content": "<[PLHD107_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 861 |
+
"lstrip": false,
|
| 862 |
+
"normalized": false,
|
| 863 |
+
"rstrip": false,
|
| 864 |
+
"single_word": false,
|
| 865 |
+
"special": true
|
| 866 |
+
},
|
| 867 |
+
"108": {
|
| 868 |
+
"content": "<[PLHD108_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 869 |
+
"lstrip": false,
|
| 870 |
+
"normalized": false,
|
| 871 |
+
"rstrip": false,
|
| 872 |
+
"single_word": false,
|
| 873 |
+
"special": true
|
| 874 |
+
},
|
| 875 |
+
"109": {
|
| 876 |
+
"content": "<[PLHD109_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 877 |
+
"lstrip": false,
|
| 878 |
+
"normalized": false,
|
| 879 |
+
"rstrip": false,
|
| 880 |
+
"single_word": false,
|
| 881 |
+
"special": true
|
| 882 |
+
},
|
| 883 |
+
"110": {
|
| 884 |
+
"content": "<[PLHD110_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 885 |
+
"lstrip": false,
|
| 886 |
+
"normalized": false,
|
| 887 |
+
"rstrip": false,
|
| 888 |
+
"single_word": false,
|
| 889 |
+
"special": true
|
| 890 |
+
},
|
| 891 |
+
"111": {
|
| 892 |
+
"content": "<[PLHD111_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 893 |
+
"lstrip": false,
|
| 894 |
+
"normalized": false,
|
| 895 |
+
"rstrip": false,
|
| 896 |
+
"single_word": false,
|
| 897 |
+
"special": true
|
| 898 |
+
},
|
| 899 |
+
"112": {
|
| 900 |
+
"content": "<[PLHD112_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 901 |
+
"lstrip": false,
|
| 902 |
+
"normalized": false,
|
| 903 |
+
"rstrip": false,
|
| 904 |
+
"single_word": false,
|
| 905 |
+
"special": true
|
| 906 |
+
},
|
| 907 |
+
"113": {
|
| 908 |
+
"content": "<[PLHD113_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 909 |
+
"lstrip": false,
|
| 910 |
+
"normalized": false,
|
| 911 |
+
"rstrip": false,
|
| 912 |
+
"single_word": false,
|
| 913 |
+
"special": true
|
| 914 |
+
},
|
| 915 |
+
"114": {
|
| 916 |
+
"content": "<[PLHD114_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 917 |
+
"lstrip": false,
|
| 918 |
+
"normalized": false,
|
| 919 |
+
"rstrip": false,
|
| 920 |
+
"single_word": false,
|
| 921 |
+
"special": true
|
| 922 |
+
},
|
| 923 |
+
"115": {
|
| 924 |
+
"content": "<[PLHD115_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 925 |
+
"lstrip": false,
|
| 926 |
+
"normalized": false,
|
| 927 |
+
"rstrip": false,
|
| 928 |
+
"single_word": false,
|
| 929 |
+
"special": true
|
| 930 |
+
},
|
| 931 |
+
"116": {
|
| 932 |
+
"content": "<[PLHD116_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 933 |
+
"lstrip": false,
|
| 934 |
+
"normalized": false,
|
| 935 |
+
"rstrip": false,
|
| 936 |
+
"single_word": false,
|
| 937 |
+
"special": true
|
| 938 |
+
},
|
| 939 |
+
"117": {
|
| 940 |
+
"content": "<[PLHD117_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 941 |
+
"lstrip": false,
|
| 942 |
+
"normalized": false,
|
| 943 |
+
"rstrip": false,
|
| 944 |
+
"single_word": false,
|
| 945 |
+
"special": true
|
| 946 |
+
},
|
| 947 |
+
"118": {
|
| 948 |
+
"content": "<[PLHD118_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 949 |
+
"lstrip": false,
|
| 950 |
+
"normalized": false,
|
| 951 |
+
"rstrip": false,
|
| 952 |
+
"single_word": false,
|
| 953 |
+
"special": true
|
| 954 |
+
},
|
| 955 |
+
"119": {
|
| 956 |
+
"content": "<[PLHD119_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 957 |
+
"lstrip": false,
|
| 958 |
+
"normalized": false,
|
| 959 |
+
"rstrip": false,
|
| 960 |
+
"single_word": false,
|
| 961 |
+
"special": true
|
| 962 |
+
},
|
| 963 |
+
"120": {
|
| 964 |
+
"content": "<[PLHD120_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 965 |
+
"lstrip": false,
|
| 966 |
+
"normalized": false,
|
| 967 |
+
"rstrip": false,
|
| 968 |
+
"single_word": false,
|
| 969 |
+
"special": true
|
| 970 |
+
},
|
| 971 |
+
"121": {
|
| 972 |
+
"content": "<[PLHD121_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 973 |
+
"lstrip": false,
|
| 974 |
+
"normalized": false,
|
| 975 |
+
"rstrip": false,
|
| 976 |
+
"single_word": false,
|
| 977 |
+
"special": true
|
| 978 |
+
},
|
| 979 |
+
"122": {
|
| 980 |
+
"content": "<[PLHD122_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 981 |
+
"lstrip": false,
|
| 982 |
+
"normalized": false,
|
| 983 |
+
"rstrip": false,
|
| 984 |
+
"single_word": false,
|
| 985 |
+
"special": true
|
| 986 |
+
},
|
| 987 |
+
"123": {
|
| 988 |
+
"content": "<[PLHD123_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 989 |
+
"lstrip": false,
|
| 990 |
+
"normalized": false,
|
| 991 |
+
"rstrip": false,
|
| 992 |
+
"single_word": false,
|
| 993 |
+
"special": true
|
| 994 |
+
},
|
| 995 |
+
"124": {
|
| 996 |
+
"content": "<[fim-prefix]>",
|
| 997 |
+
"lstrip": false,
|
| 998 |
+
"normalized": false,
|
| 999 |
+
"rstrip": false,
|
| 1000 |
+
"single_word": false,
|
| 1001 |
+
"special": true
|
| 1002 |
+
},
|
| 1003 |
+
"125": {
|
| 1004 |
+
"content": "<[fim-suffix]>",
|
| 1005 |
+
"lstrip": false,
|
| 1006 |
+
"normalized": false,
|
| 1007 |
+
"rstrip": false,
|
| 1008 |
+
"single_word": false,
|
| 1009 |
+
"special": true
|
| 1010 |
+
},
|
| 1011 |
+
"126": {
|
| 1012 |
+
"content": "<[fim-middle]>",
|
| 1013 |
+
"lstrip": false,
|
| 1014 |
+
"normalized": false,
|
| 1015 |
+
"rstrip": false,
|
| 1016 |
+
"single_word": false,
|
| 1017 |
+
"special": true
|
| 1018 |
+
},
|
| 1019 |
+
"127": {
|
| 1020 |
+
"content": "<[PLHD127_never_used_51bce0c785ca2f68081bfa7d91973934]>",
|
| 1021 |
+
"lstrip": false,
|
| 1022 |
+
"normalized": false,
|
| 1023 |
+
"rstrip": false,
|
| 1024 |
+
"single_word": false,
|
| 1025 |
+
"special": true
|
| 1026 |
+
}
|
| 1027 |
+
},
|
| 1028 |
+
"bos_token": "<[begin▁of▁sentence]>",
|
| 1029 |
+
"chat_template": "{% set has_system = messages[0]['role'] == 'system' %}{% if not has_system %}{{ bos_token + 'system\nYou are an AI programming assistant, utilizing the Seed-Coder model, developed by ByteDance Seed, and you only answer questions related to computer science. For politically sensitive questions, security and privacy issues, and other non-computer science questions, you will refuse to answer.\n\n' + eos_token }}{% endif %}{% for message in messages %}{{ bos_token + message['role'] + '\n' + message['content'] | trim + eos_token }}{% endfor %}{% if add_generation_prompt %}{{ bos_token + 'assistant\n'}}{% endif %}",
|
| 1030 |
+
"clean_up_tokenization_spaces": false,
|
| 1031 |
+
"eos_token": "<[end▁of▁sentence]>",
|
| 1032 |
+
"extra_special_tokens": {},
|
| 1033 |
+
"model_max_length": 32768,
|
| 1034 |
+
"pad_token": "<[PAD▁TOKEN]>",
|
| 1035 |
+
"sep_token": "<[SEP▁TOKEN]>",
|
| 1036 |
+
"tokenizer_class": "PreTrainedTokenizer"
|
| 1037 |
+
}
|