Upload 12 files
Browse files- .gitattributes +1 -0
- added_tokens.json +3 -0
- config.json +58 -0
- generation_config.json +11 -0
- model-00001-of-00002.safetensors +3 -0
- model-00002-of-00002.safetensors +3 -0
- model.safetensors.index.json +0 -0
- quant_log.csv +337 -0
- quantize_config.json +21 -0
- special_tokens_map.json +27 -0
- tokenizer.json +3 -0
- tokenizer.model +3 -0
- tokenizer_config.json +0 -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
|
added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<image_soft_token>": 262144
|
| 3 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Gemma3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"attn_logit_softcapping": null,
|
| 8 |
+
"bos_token_id": 2,
|
| 9 |
+
"cache_implementation": "hybrid",
|
| 10 |
+
"eos_token_id": 1,
|
| 11 |
+
"final_logit_softcapping": null,
|
| 12 |
+
"head_dim": 256,
|
| 13 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 14 |
+
"hidden_size": 3840,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 15360,
|
| 17 |
+
"max_position_embeddings": 131072,
|
| 18 |
+
"model_type": "gemma3_text",
|
| 19 |
+
"num_attention_heads": 16,
|
| 20 |
+
"num_hidden_layers": 48,
|
| 21 |
+
"num_key_value_heads": 8,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"quantization_config": {
|
| 24 |
+
"bits": 4,
|
| 25 |
+
"checkpoint_format": "gptq",
|
| 26 |
+
"desc_act": true,
|
| 27 |
+
"group_size": 128,
|
| 28 |
+
"lm_head": false,
|
| 29 |
+
"meta": {
|
| 30 |
+
"damp_auto_increment": 0.0025,
|
| 31 |
+
"damp_percent": 0.01,
|
| 32 |
+
"mse": 0.0,
|
| 33 |
+
"quantizer": [
|
| 34 |
+
"gptqmodel:2.1.0"
|
| 35 |
+
],
|
| 36 |
+
"static_groups": false,
|
| 37 |
+
"true_sequential": true,
|
| 38 |
+
"uri": "https://github.com/modelcloud/gptqmodel"
|
| 39 |
+
},
|
| 40 |
+
"pack_dtype": "int32",
|
| 41 |
+
"quant_method": "gptq",
|
| 42 |
+
"sym": true
|
| 43 |
+
},
|
| 44 |
+
"query_pre_attn_scalar": 256,
|
| 45 |
+
"rms_norm_eps": 1e-06,
|
| 46 |
+
"rope_local_base_freq": 10000.0,
|
| 47 |
+
"rope_scaling": {
|
| 48 |
+
"factor": 8.0,
|
| 49 |
+
"rope_type": "linear"
|
| 50 |
+
},
|
| 51 |
+
"rope_theta": 1000000.0,
|
| 52 |
+
"sliding_window": 1024,
|
| 53 |
+
"sliding_window_pattern": 6,
|
| 54 |
+
"torch_dtype": "bfloat16",
|
| 55 |
+
"transformers_version": "4.51.0.dev0",
|
| 56 |
+
"use_cache": true,
|
| 57 |
+
"vocab_size": 262208
|
| 58 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"cache_implementation": "hybrid",
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
1,
|
| 7 |
+
106
|
| 8 |
+
],
|
| 9 |
+
"pad_token_id": 0,
|
| 10 |
+
"transformers_version": "4.51.0.dev0"
|
| 11 |
+
}
|
model-00001-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:002703dba69e24b19003d2b5c8e0a488531071e5fb581388678e18f2be3572f1
|
| 3 |
+
size 3996541560
|
model-00002-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:48a5ca364000adead0d565a11f7474446c55cf83d54c73ecb7fba98230c104f8
|
| 3 |
+
size 3615673656
|
model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
quant_log.csv
ADDED
|
@@ -0,0 +1,337 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
layer,module,loss,damp,time
|
| 2 |
+
0,self_attn.k_proj,541.13354,0.01000,1.369
|
| 3 |
+
0,self_attn.v_proj,524.00024,0.01000,1.146
|
| 4 |
+
0,self_attn.q_proj,236.65874,0.01000,1.159
|
| 5 |
+
0,self_attn.o_proj,313.03809,0.01000,1.239
|
| 6 |
+
0,mlp.up_proj,13.09483,0.01000,1.176
|
| 7 |
+
0,mlp.gate_proj,13.77142,0.01000,1.179
|
| 8 |
+
0,mlp.down_proj,4.23081,0.01000,5.288
|
| 9 |
+
1,self_attn.k_proj,17.93688,0.01000,1.146
|
| 10 |
+
1,self_attn.v_proj,18.72330,0.01000,1.141
|
| 11 |
+
1,self_attn.q_proj,19.96606,0.01000,1.168
|
| 12 |
+
1,self_attn.o_proj,2.47787,0.01000,1.230
|
| 13 |
+
1,mlp.up_proj,31.92973,0.01000,1.177
|
| 14 |
+
1,mlp.gate_proj,43.11442,0.01000,1.171
|
| 15 |
+
1,mlp.down_proj,2.11448,0.01000,5.270
|
| 16 |
+
2,self_attn.k_proj,55.17967,0.01000,1.140
|
| 17 |
+
2,self_attn.v_proj,56.39925,0.01000,1.138
|
| 18 |
+
2,self_attn.q_proj,20.52049,0.01000,1.153
|
| 19 |
+
2,self_attn.o_proj,16.94455,0.01000,1.224
|
| 20 |
+
2,mlp.up_proj,26.67284,0.01000,1.173
|
| 21 |
+
2,mlp.gate_proj,29.84965,0.01000,1.166
|
| 22 |
+
2,mlp.down_proj,0.43737,0.01000,5.264
|
| 23 |
+
3,self_attn.k_proj,32.86499,0.01000,1.140
|
| 24 |
+
3,self_attn.v_proj,33.66631,0.01000,1.145
|
| 25 |
+
3,self_attn.q_proj,21.27360,0.01000,1.162
|
| 26 |
+
3,self_attn.o_proj,15.54027,0.01000,1.232
|
| 27 |
+
3,mlp.up_proj,175.99846,0.01000,1.166
|
| 28 |
+
3,mlp.gate_proj,183.83678,0.01000,1.168
|
| 29 |
+
3,mlp.down_proj,43.10832,0.01000,5.278
|
| 30 |
+
4,self_attn.k_proj,50.42262,0.01000,1.134
|
| 31 |
+
4,self_attn.v_proj,50.80437,0.01000,1.128
|
| 32 |
+
4,self_attn.q_proj,34.96879,0.01000,1.160
|
| 33 |
+
4,self_attn.o_proj,20.64001,0.01000,1.227
|
| 34 |
+
4,mlp.up_proj,42.46541,0.01000,1.170
|
| 35 |
+
4,mlp.gate_proj,44.23412,0.01000,1.165
|
| 36 |
+
4,mlp.down_proj,1.56396,0.01000,5.256
|
| 37 |
+
5,self_attn.k_proj,72.72899,0.01000,1.142
|
| 38 |
+
5,self_attn.v_proj,58.08025,0.01000,1.134
|
| 39 |
+
5,self_attn.q_proj,17.82669,0.01000,1.166
|
| 40 |
+
5,self_attn.o_proj,5.53370,0.01000,1.225
|
| 41 |
+
5,mlp.up_proj,42.33241,0.01000,1.172
|
| 42 |
+
5,mlp.gate_proj,44.19517,0.01000,1.169
|
| 43 |
+
5,mlp.down_proj,2.22423,0.01000,5.276
|
| 44 |
+
6,self_attn.k_proj,84.36617,0.01000,1.141
|
| 45 |
+
6,self_attn.v_proj,68.58587,0.01000,1.141
|
| 46 |
+
6,self_attn.q_proj,42.22988,0.01000,1.165
|
| 47 |
+
6,self_attn.o_proj,22.37763,0.01000,1.224
|
| 48 |
+
6,mlp.up_proj,22.91920,0.01000,1.165
|
| 49 |
+
6,mlp.gate_proj,23.08118,0.01000,1.176
|
| 50 |
+
6,mlp.down_proj,0.41280,0.01000,5.278
|
| 51 |
+
7,self_attn.k_proj,75.11481,0.01000,1.146
|
| 52 |
+
7,self_attn.v_proj,69.92151,0.01000,1.136
|
| 53 |
+
7,self_attn.q_proj,23.31736,0.01000,1.165
|
| 54 |
+
7,self_attn.o_proj,17.73646,0.01000,1.228
|
| 55 |
+
7,mlp.up_proj,53.84304,0.01000,1.175
|
| 56 |
+
7,mlp.gate_proj,54.73155,0.01000,1.168
|
| 57 |
+
7,mlp.down_proj,1.29550,0.01000,5.276
|
| 58 |
+
8,self_attn.k_proj,29.92232,0.01000,1.130
|
| 59 |
+
8,self_attn.v_proj,25.64258,0.01000,1.135
|
| 60 |
+
8,self_attn.q_proj,49.92485,0.01000,1.155
|
| 61 |
+
8,self_attn.o_proj,4.97905,0.01000,1.226
|
| 62 |
+
8,mlp.up_proj,41.39110,0.01000,1.165
|
| 63 |
+
8,mlp.gate_proj,41.33935,0.01000,1.166
|
| 64 |
+
8,mlp.down_proj,0.40674,0.01000,5.255
|
| 65 |
+
9,self_attn.k_proj,45.50682,0.01000,1.143
|
| 66 |
+
9,self_attn.v_proj,43.90495,0.01000,1.157
|
| 67 |
+
9,self_attn.q_proj,7.16775,0.01000,1.205
|
| 68 |
+
9,self_attn.o_proj,8.32300,0.01000,1.255
|
| 69 |
+
9,mlp.up_proj,18.17541,0.01000,1.190
|
| 70 |
+
9,mlp.gate_proj,18.48425,0.01000,1.185
|
| 71 |
+
9,mlp.down_proj,0.32448,0.01000,5.336
|
| 72 |
+
10,self_attn.k_proj,30.81048,0.01000,1.152
|
| 73 |
+
10,self_attn.v_proj,32.14077,0.01000,1.144
|
| 74 |
+
10,self_attn.q_proj,11.03526,0.01000,1.173
|
| 75 |
+
10,self_attn.o_proj,6.05486,0.01000,1.231
|
| 76 |
+
10,mlp.up_proj,19.89635,0.01000,1.175
|
| 77 |
+
10,mlp.gate_proj,20.11558,0.01000,1.179
|
| 78 |
+
10,mlp.down_proj,0.45918,0.01000,5.307
|
| 79 |
+
11,self_attn.k_proj,36.07178,0.01000,1.157
|
| 80 |
+
11,self_attn.v_proj,23.75871,0.01000,1.155
|
| 81 |
+
11,self_attn.q_proj,4.07136,0.01000,1.174
|
| 82 |
+
11,self_attn.o_proj,4.88146,0.01000,1.246
|
| 83 |
+
11,mlp.up_proj,16.78284,0.01000,1.191
|
| 84 |
+
11,mlp.gate_proj,16.65389,0.01000,1.181
|
| 85 |
+
11,mlp.down_proj,0.22747,0.01000,5.326
|
| 86 |
+
12,self_attn.k_proj,28.23595,0.01000,1.151
|
| 87 |
+
12,self_attn.v_proj,24.78283,0.01000,1.146
|
| 88 |
+
12,self_attn.q_proj,6.14941,0.01000,1.169
|
| 89 |
+
12,self_attn.o_proj,2.96259,0.01000,1.233
|
| 90 |
+
12,mlp.up_proj,19.28688,0.01000,1.183
|
| 91 |
+
12,mlp.gate_proj,19.24973,0.01000,1.179
|
| 92 |
+
12,mlp.down_proj,0.43755,0.01000,5.330
|
| 93 |
+
13,self_attn.k_proj,28.39410,0.01000,1.158
|
| 94 |
+
13,self_attn.v_proj,27.71966,0.01000,1.152
|
| 95 |
+
13,self_attn.q_proj,5.89212,0.01000,1.176
|
| 96 |
+
13,self_attn.o_proj,2.00516,0.01000,1.265
|
| 97 |
+
13,mlp.up_proj,13.19166,0.01000,1.196
|
| 98 |
+
13,mlp.gate_proj,13.18062,0.01000,1.214
|
| 99 |
+
13,mlp.down_proj,0.26906,0.01000,5.319
|
| 100 |
+
14,self_attn.k_proj,5.90705,0.01000,1.153
|
| 101 |
+
14,self_attn.v_proj,5.37098,0.01000,1.151
|
| 102 |
+
14,self_attn.q_proj,8.23420,0.01000,1.176
|
| 103 |
+
14,self_attn.o_proj,1.25329,0.01000,1.233
|
| 104 |
+
14,mlp.up_proj,5.87847,0.01000,1.180
|
| 105 |
+
14,mlp.gate_proj,5.87578,0.01000,1.167
|
| 106 |
+
14,mlp.down_proj,0.44269,0.01000,5.268
|
| 107 |
+
15,self_attn.k_proj,3.62104,0.01000,1.159
|
| 108 |
+
15,self_attn.v_proj,3.31765,0.01000,1.164
|
| 109 |
+
15,self_attn.q_proj,6.65585,0.01000,1.185
|
| 110 |
+
15,self_attn.o_proj,1.09457,0.01000,1.249
|
| 111 |
+
15,mlp.up_proj,5.64707,0.01000,1.187
|
| 112 |
+
15,mlp.gate_proj,5.64888,0.01000,1.183
|
| 113 |
+
15,mlp.down_proj,1.79231,0.01000,5.323
|
| 114 |
+
16,self_attn.k_proj,5.06185,0.01000,1.158
|
| 115 |
+
16,self_attn.v_proj,5.04160,0.01000,1.156
|
| 116 |
+
16,self_attn.q_proj,2.51695,0.01000,1.182
|
| 117 |
+
16,self_attn.o_proj,1.42447,0.01000,1.238
|
| 118 |
+
16,mlp.up_proj,4.45154,0.01000,1.184
|
| 119 |
+
16,mlp.gate_proj,4.43117,0.01000,1.185
|
| 120 |
+
16,mlp.down_proj,0.10553,0.01000,5.312
|
| 121 |
+
17,self_attn.k_proj,1.61782,0.01000,1.157
|
| 122 |
+
17,self_attn.v_proj,1.11016,0.01000,1.154
|
| 123 |
+
17,self_attn.q_proj,1.51316,0.01000,1.183
|
| 124 |
+
17,self_attn.o_proj,0.21809,0.01000,1.249
|
| 125 |
+
17,mlp.up_proj,10.56579,0.01000,1.203
|
| 126 |
+
17,mlp.gate_proj,10.49751,0.01000,1.197
|
| 127 |
+
17,mlp.down_proj,0.07942,0.01000,5.333
|
| 128 |
+
18,self_attn.k_proj,1.67368,0.01000,1.153
|
| 129 |
+
18,self_attn.v_proj,1.47577,0.01000,1.150
|
| 130 |
+
18,self_attn.q_proj,1.22046,0.01000,1.173
|
| 131 |
+
18,self_attn.o_proj,0.25480,0.01000,1.236
|
| 132 |
+
18,mlp.up_proj,6.13185,0.01000,1.177
|
| 133 |
+
18,mlp.gate_proj,6.06935,0.01000,1.180
|
| 134 |
+
18,mlp.down_proj,0.34597,0.01000,5.333
|
| 135 |
+
19,self_attn.k_proj,4.36240,0.01000,1.163
|
| 136 |
+
19,self_attn.v_proj,3.68388,0.01000,1.156
|
| 137 |
+
19,self_attn.q_proj,1.44465,0.01000,1.181
|
| 138 |
+
19,self_attn.o_proj,0.76993,0.01000,1.244
|
| 139 |
+
19,mlp.up_proj,3.91250,0.01000,1.173
|
| 140 |
+
19,mlp.gate_proj,3.84983,0.01000,1.172
|
| 141 |
+
19,mlp.down_proj,0.07436,0.01000,5.274
|
| 142 |
+
20,self_attn.k_proj,8.00628,0.01000,1.147
|
| 143 |
+
20,self_attn.v_proj,7.67947,0.01000,1.145
|
| 144 |
+
20,self_attn.q_proj,2.27584,0.01000,1.166
|
| 145 |
+
20,self_attn.o_proj,0.54385,0.01000,1.231
|
| 146 |
+
20,mlp.up_proj,4.61316,0.01000,1.178
|
| 147 |
+
20,mlp.gate_proj,4.56322,0.01000,1.177
|
| 148 |
+
20,mlp.down_proj,0.14599,0.01000,5.310
|
| 149 |
+
21,self_attn.k_proj,2.59709,0.01000,1.136
|
| 150 |
+
21,self_attn.v_proj,2.45972,0.01000,1.137
|
| 151 |
+
21,self_attn.q_proj,4.14215,0.01000,1.167
|
| 152 |
+
21,self_attn.o_proj,0.32877,0.01000,1.235
|
| 153 |
+
21,mlp.up_proj,5.49134,0.01000,1.175
|
| 154 |
+
21,mlp.gate_proj,5.51805,0.01000,1.178
|
| 155 |
+
21,mlp.down_proj,0.16652,0.01000,5.256
|
| 156 |
+
22,self_attn.k_proj,3.19467,0.01000,1.135
|
| 157 |
+
22,self_attn.v_proj,2.79026,0.01000,1.141
|
| 158 |
+
22,self_attn.q_proj,1.95016,0.01000,1.161
|
| 159 |
+
22,self_attn.o_proj,0.62331,0.01000,1.228
|
| 160 |
+
22,mlp.up_proj,4.84731,0.01000,1.178
|
| 161 |
+
22,mlp.gate_proj,4.81332,0.01000,1.171
|
| 162 |
+
22,mlp.down_proj,0.02194,0.01000,5.270
|
| 163 |
+
23,self_attn.k_proj,1.24092,0.01000,1.144
|
| 164 |
+
23,self_attn.v_proj,0.96560,0.01000,1.174
|
| 165 |
+
23,self_attn.q_proj,0.69119,0.01000,1.209
|
| 166 |
+
23,self_attn.o_proj,0.05925,0.01000,1.266
|
| 167 |
+
23,mlp.up_proj,4.66686,0.01000,1.182
|
| 168 |
+
23,mlp.gate_proj,4.78342,0.01000,1.179
|
| 169 |
+
23,mlp.down_proj,0.03919,0.01000,5.320
|
| 170 |
+
24,self_attn.k_proj,15.45271,0.01000,1.152
|
| 171 |
+
24,self_attn.v_proj,13.04742,0.01000,1.153
|
| 172 |
+
24,self_attn.q_proj,3.05512,0.01000,1.175
|
| 173 |
+
24,self_attn.o_proj,0.56290,0.01000,1.239
|
| 174 |
+
24,mlp.up_proj,3.25264,0.01000,1.178
|
| 175 |
+
24,mlp.gate_proj,3.27618,0.01000,1.177
|
| 176 |
+
24,mlp.down_proj,0.02530,0.01000,5.320
|
| 177 |
+
25,self_attn.k_proj,3.83113,0.01000,1.149
|
| 178 |
+
25,self_attn.v_proj,3.51526,0.01000,1.148
|
| 179 |
+
25,self_attn.q_proj,6.40574,0.01000,1.172
|
| 180 |
+
25,self_attn.o_proj,0.31738,0.01000,1.232
|
| 181 |
+
25,mlp.up_proj,3.27430,0.01000,1.181
|
| 182 |
+
25,mlp.gate_proj,3.36226,0.01000,1.179
|
| 183 |
+
25,mlp.down_proj,0.00939,0.01000,5.350
|
| 184 |
+
26,self_attn.k_proj,6.09987,0.01000,1.179
|
| 185 |
+
26,self_attn.v_proj,5.23842,0.01000,1.177
|
| 186 |
+
26,self_attn.q_proj,5.84130,0.01000,1.180
|
| 187 |
+
26,self_attn.o_proj,0.83060,0.01000,1.283
|
| 188 |
+
26,mlp.up_proj,3.63879,0.01000,1.181
|
| 189 |
+
26,mlp.gate_proj,3.80557,0.01000,1.208
|
| 190 |
+
26,mlp.down_proj,0.02998,0.01000,5.355
|
| 191 |
+
27,self_attn.k_proj,6.89362,0.01000,1.148
|
| 192 |
+
27,self_attn.v_proj,5.93298,0.01000,1.152
|
| 193 |
+
27,self_attn.q_proj,4.85018,0.01000,1.181
|
| 194 |
+
27,self_attn.o_proj,0.29430,0.01000,1.240
|
| 195 |
+
27,mlp.up_proj,4.16985,0.01000,1.182
|
| 196 |
+
27,mlp.gate_proj,4.32480,0.01000,1.174
|
| 197 |
+
27,mlp.down_proj,0.01340,0.01000,5.319
|
| 198 |
+
28,self_attn.k_proj,13.89592,0.01000,1.182
|
| 199 |
+
28,self_attn.v_proj,13.21770,0.01000,1.185
|
| 200 |
+
28,self_attn.q_proj,7.81782,0.01000,1.186
|
| 201 |
+
28,self_attn.o_proj,0.88097,0.01000,1.234
|
| 202 |
+
28,mlp.up_proj,15.60182,0.01000,1.182
|
| 203 |
+
28,mlp.gate_proj,16.38438,0.01000,1.205
|
| 204 |
+
28,mlp.down_proj,0.13926,0.01000,5.322
|
| 205 |
+
29,self_attn.k_proj,3.39715,0.01000,1.156
|
| 206 |
+
29,self_attn.v_proj,3.12082,0.01000,1.159
|
| 207 |
+
29,self_attn.q_proj,2.24113,0.01000,1.178
|
| 208 |
+
29,self_attn.o_proj,0.26442,0.01000,1.237
|
| 209 |
+
29,mlp.up_proj,13.86748,0.01000,1.186
|
| 210 |
+
29,mlp.gate_proj,14.45333,0.01000,1.186
|
| 211 |
+
29,mlp.down_proj,0.12276,0.01000,5.333
|
| 212 |
+
30,self_attn.k_proj,8.86315,0.01000,1.151
|
| 213 |
+
30,self_attn.v_proj,9.34150,0.01000,1.149
|
| 214 |
+
30,self_attn.q_proj,15.43161,0.01000,1.170
|
| 215 |
+
30,self_attn.o_proj,0.52530,0.01000,1.251
|
| 216 |
+
30,mlp.up_proj,21.33563,0.01000,1.203
|
| 217 |
+
30,mlp.gate_proj,21.95199,0.01000,1.206
|
| 218 |
+
30,mlp.down_proj,0.33765,0.01000,5.337
|
| 219 |
+
31,self_attn.k_proj,17.89855,0.01000,1.161
|
| 220 |
+
31,self_attn.v_proj,15.04204,0.01000,1.155
|
| 221 |
+
31,self_attn.q_proj,24.59508,0.01000,1.176
|
| 222 |
+
31,self_attn.o_proj,1.00914,0.01000,1.244
|
| 223 |
+
31,mlp.up_proj,10.30728,0.01000,1.183
|
| 224 |
+
31,mlp.gate_proj,10.62114,0.01000,1.184
|
| 225 |
+
31,mlp.down_proj,0.10517,0.01000,5.316
|
| 226 |
+
32,self_attn.k_proj,13.39213,0.01000,1.153
|
| 227 |
+
32,self_attn.v_proj,12.69509,0.01000,1.154
|
| 228 |
+
32,self_attn.q_proj,21.92416,0.01000,1.180
|
| 229 |
+
32,self_attn.o_proj,1.35059,0.01000,1.235
|
| 230 |
+
32,mlp.up_proj,20.39222,0.01000,1.222
|
| 231 |
+
32,mlp.gate_proj,21.17166,0.01000,1.223
|
| 232 |
+
32,mlp.down_proj,0.14895,0.01000,5.467
|
| 233 |
+
33,self_attn.k_proj,67.60665,0.01000,1.158
|
| 234 |
+
33,self_attn.v_proj,61.93232,0.01000,1.156
|
| 235 |
+
33,self_attn.q_proj,10.20906,0.01000,1.170
|
| 236 |
+
33,self_attn.o_proj,6.49278,0.01000,1.237
|
| 237 |
+
33,mlp.up_proj,21.61745,0.01000,1.188
|
| 238 |
+
33,mlp.gate_proj,22.45165,0.01000,1.183
|
| 239 |
+
33,mlp.down_proj,0.40881,0.01000,5.329
|
| 240 |
+
34,self_attn.k_proj,25.21844,0.01000,1.150
|
| 241 |
+
34,self_attn.v_proj,23.25540,0.01000,1.149
|
| 242 |
+
34,self_attn.q_proj,23.43749,0.01000,1.173
|
| 243 |
+
34,self_attn.o_proj,0.89653,0.01000,1.238
|
| 244 |
+
34,mlp.up_proj,23.62345,0.01000,1.181
|
| 245 |
+
34,mlp.gate_proj,24.54617,0.01000,1.177
|
| 246 |
+
34,mlp.down_proj,0.47707,0.01000,5.344
|
| 247 |
+
35,self_attn.k_proj,4.05301,0.01000,1.148
|
| 248 |
+
35,self_attn.v_proj,3.72610,0.01000,1.152
|
| 249 |
+
35,self_attn.q_proj,5.03022,0.01000,1.177
|
| 250 |
+
35,self_attn.o_proj,0.15661,0.01000,1.238
|
| 251 |
+
35,mlp.up_proj,24.17026,0.01000,1.181
|
| 252 |
+
35,mlp.gate_proj,24.90578,0.01000,1.188
|
| 253 |
+
35,mlp.down_proj,0.28311,0.01000,5.365
|
| 254 |
+
36,self_attn.k_proj,20.25990,0.01000,1.158
|
| 255 |
+
36,self_attn.v_proj,17.49859,0.01000,1.148
|
| 256 |
+
36,self_attn.q_proj,36.25520,0.01000,1.184
|
| 257 |
+
36,self_attn.o_proj,1.39140,0.01000,1.236
|
| 258 |
+
36,mlp.up_proj,24.74201,0.01000,1.178
|
| 259 |
+
36,mlp.gate_proj,25.37830,0.01000,1.174
|
| 260 |
+
36,mlp.down_proj,0.17330,0.01000,5.284
|
| 261 |
+
37,self_attn.k_proj,56.08345,0.01000,1.155
|
| 262 |
+
37,self_attn.v_proj,52.79922,0.01000,1.147
|
| 263 |
+
37,self_attn.q_proj,33.67844,0.01000,1.168
|
| 264 |
+
37,self_attn.o_proj,0.92829,0.01000,1.236
|
| 265 |
+
37,mlp.up_proj,18.60583,0.01000,1.174
|
| 266 |
+
37,mlp.gate_proj,18.98741,0.01000,1.177
|
| 267 |
+
37,mlp.down_proj,0.09705,0.01000,5.314
|
| 268 |
+
38,self_attn.k_proj,41.53551,0.01000,1.143
|
| 269 |
+
38,self_attn.v_proj,33.07961,0.01000,1.141
|
| 270 |
+
38,self_attn.q_proj,26.74597,0.01000,1.167
|
| 271 |
+
38,self_attn.o_proj,1.16061,0.01000,1.229
|
| 272 |
+
38,mlp.up_proj,16.90224,0.01000,1.176
|
| 273 |
+
38,mlp.gate_proj,17.23815,0.01000,1.178
|
| 274 |
+
38,mlp.down_proj,0.25427,0.01000,5.282
|
| 275 |
+
39,self_attn.k_proj,17.96823,0.01000,1.155
|
| 276 |
+
39,self_attn.v_proj,17.67106,0.01000,1.144
|
| 277 |
+
39,self_attn.q_proj,33.80338,0.01000,1.170
|
| 278 |
+
39,self_attn.o_proj,0.67802,0.01000,1.234
|
| 279 |
+
39,mlp.up_proj,19.49164,0.01000,1.172
|
| 280 |
+
39,mlp.gate_proj,19.95888,0.01000,1.170
|
| 281 |
+
39,mlp.down_proj,0.37972,0.01000,5.311
|
| 282 |
+
40,self_attn.k_proj,38.02741,0.01000,1.154
|
| 283 |
+
40,self_attn.v_proj,36.83753,0.01000,1.151
|
| 284 |
+
40,self_attn.q_proj,28.99961,0.01000,1.182
|
| 285 |
+
40,self_attn.o_proj,0.60003,0.01000,1.247
|
| 286 |
+
40,mlp.up_proj,16.13040,0.01000,1.187
|
| 287 |
+
40,mlp.gate_proj,16.71348,0.01000,1.184
|
| 288 |
+
40,mlp.down_proj,0.13565,0.01000,5.350
|
| 289 |
+
41,self_attn.k_proj,3.06665,0.01000,1.153
|
| 290 |
+
41,self_attn.v_proj,2.25483,0.01000,1.146
|
| 291 |
+
41,self_attn.q_proj,1.40651,0.01000,1.176
|
| 292 |
+
41,self_attn.o_proj,0.02935,0.01000,1.247
|
| 293 |
+
41,mlp.up_proj,21.41223,0.01000,1.187
|
| 294 |
+
41,mlp.gate_proj,22.10480,0.01000,1.179
|
| 295 |
+
41,mlp.down_proj,0.32669,0.01000,5.312
|
| 296 |
+
42,self_attn.k_proj,12.04000,0.01000,1.147
|
| 297 |
+
42,self_attn.v_proj,9.32724,0.01000,1.146
|
| 298 |
+
42,self_attn.q_proj,17.59069,0.01000,1.165
|
| 299 |
+
42,self_attn.o_proj,0.25627,0.01000,1.239
|
| 300 |
+
42,mlp.up_proj,10.66330,0.01000,1.175
|
| 301 |
+
42,mlp.gate_proj,11.14412,0.01000,1.173
|
| 302 |
+
42,mlp.down_proj,0.11307,0.01000,5.324
|
| 303 |
+
43,self_attn.k_proj,19.24364,0.01000,1.149
|
| 304 |
+
43,self_attn.v_proj,16.14411,0.01000,1.144
|
| 305 |
+
43,self_attn.q_proj,2.13901,0.01000,1.171
|
| 306 |
+
43,self_attn.o_proj,0.57865,0.01000,1.236
|
| 307 |
+
43,mlp.up_proj,17.27629,0.01000,1.185
|
| 308 |
+
43,mlp.gate_proj,18.36621,0.01000,1.176
|
| 309 |
+
43,mlp.down_proj,0.19911,0.01000,5.313
|
| 310 |
+
44,self_attn.k_proj,56.45793,0.01000,1.155
|
| 311 |
+
44,self_attn.v_proj,42.57193,0.01000,1.155
|
| 312 |
+
44,self_attn.q_proj,9.74599,0.01000,1.174
|
| 313 |
+
44,self_attn.o_proj,4.25844,0.01000,1.232
|
| 314 |
+
44,mlp.up_proj,12.01761,0.01000,1.178
|
| 315 |
+
44,mlp.gate_proj,12.94910,0.01000,1.172
|
| 316 |
+
44,mlp.down_proj,0.22308,0.01000,5.293
|
| 317 |
+
45,self_attn.k_proj,33.53449,0.01000,1.151
|
| 318 |
+
45,self_attn.v_proj,27.88224,0.01000,1.146
|
| 319 |
+
45,self_attn.q_proj,25.26192,0.01000,1.170
|
| 320 |
+
45,self_attn.o_proj,1.09609,0.01000,1.233
|
| 321 |
+
45,mlp.up_proj,12.26534,0.01000,1.181
|
| 322 |
+
45,mlp.gate_proj,13.05515,0.01000,1.180
|
| 323 |
+
45,mlp.down_proj,0.31663,0.01000,5.342
|
| 324 |
+
46,self_attn.k_proj,21.10816,0.01000,1.150
|
| 325 |
+
46,self_attn.v_proj,19.91315,0.01000,1.144
|
| 326 |
+
46,self_attn.q_proj,3.43455,0.01000,1.165
|
| 327 |
+
46,self_attn.o_proj,1.93408,0.01000,1.237
|
| 328 |
+
46,mlp.up_proj,30.78687,0.01000,1.180
|
| 329 |
+
46,mlp.gate_proj,31.95230,0.01000,1.176
|
| 330 |
+
46,mlp.down_proj,0.45266,0.01000,5.307
|
| 331 |
+
47,self_attn.k_proj,2.52750,0.01000,1.155
|
| 332 |
+
47,self_attn.v_proj,2.22451,0.01000,1.149
|
| 333 |
+
47,self_attn.q_proj,2.76382,0.01000,1.170
|
| 334 |
+
47,self_attn.o_proj,0.18812,0.01000,1.245
|
| 335 |
+
47,mlp.up_proj,26.70221,0.01000,1.177
|
| 336 |
+
47,mlp.gate_proj,27.67697,0.01000,1.178
|
| 337 |
+
47,mlp.down_proj,0.65549,0.01000,5.359
|
quantize_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bits": 4,
|
| 3 |
+
"group_size": 128,
|
| 4 |
+
"desc_act": true,
|
| 5 |
+
"sym": true,
|
| 6 |
+
"lm_head": false,
|
| 7 |
+
"quant_method": "gptq",
|
| 8 |
+
"checkpoint_format": "gptq",
|
| 9 |
+
"pack_dtype": "int32",
|
| 10 |
+
"meta": {
|
| 11 |
+
"quantizer": [
|
| 12 |
+
"gptqmodel:2.1.0"
|
| 13 |
+
],
|
| 14 |
+
"uri": "https://github.com/modelcloud/gptqmodel",
|
| 15 |
+
"damp_percent": 0.01,
|
| 16 |
+
"damp_auto_increment": 0.0025,
|
| 17 |
+
"static_groups": false,
|
| 18 |
+
"true_sequential": true,
|
| 19 |
+
"mse": 0.0
|
| 20 |
+
}
|
| 21 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"boi_token": "<start_of_image>",
|
| 3 |
+
"bos_token": {
|
| 4 |
+
"content": "<bos>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
"eoi_token": "<end_of_image>",
|
| 11 |
+
"eos_token": {
|
| 12 |
+
"content": "<end_of_turn>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false
|
| 17 |
+
},
|
| 18 |
+
"image_token": "<image_soft_token>",
|
| 19 |
+
"pad_token": "<pad>",
|
| 20 |
+
"unk_token": {
|
| 21 |
+
"content": "<unk>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false
|
| 26 |
+
}
|
| 27 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
|
| 3 |
+
size 33384568
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
|
| 3 |
+
size 4689074
|
tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|