Upload GPTQ quantized model
Browse files- .gitattributes +1 -0
- chat_template.jinja +1 -0
- config.json +193 -0
- generation_config.json +11 -0
- model.safetensors +3 -0
- quant_log.csv +161 -0
- quantize_config.json +47 -0
- tokenizer.json +3 -0
- tokenizer_config.json +13 -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
|
chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}{% if message['role'] == 'system' and 'tools' in message and message['tools'] is not none %}{{ '<|' + message['role'] + '|>' + message['content'] + '<|tool|>' + message['tools'] + '<|/tool|>' + '<|end|>' }}{% else %}{{ '<|' + message['role'] + '|>' + message['content'] + '<|end|>' }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>' }}{% else %}{{ eos_token }}{% endif %}
|
config.json
ADDED
|
@@ -0,0 +1,193 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Phi3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"auto_map": {
|
| 8 |
+
"AutoConfig": "configuration_phi3.Phi3Config",
|
| 9 |
+
"AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM",
|
| 10 |
+
"AutoTokenizer": "Xenova/gpt-4o"
|
| 11 |
+
},
|
| 12 |
+
"bos_token_id": 199999,
|
| 13 |
+
"dtype": "bfloat16",
|
| 14 |
+
"embd_pdrop": 0.0,
|
| 15 |
+
"eos_token_id": 199999,
|
| 16 |
+
"full_attn_mod": 1,
|
| 17 |
+
"hidden_act": "silu",
|
| 18 |
+
"hidden_size": 3072,
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 8192,
|
| 21 |
+
"interpolate_factor": 1,
|
| 22 |
+
"lm_head_bias": false,
|
| 23 |
+
"max_position_embeddings": 131072,
|
| 24 |
+
"mlp_bias": false,
|
| 25 |
+
"model_type": "phi3",
|
| 26 |
+
"num_attention_heads": 24,
|
| 27 |
+
"num_hidden_layers": 32,
|
| 28 |
+
"num_key_value_heads": 8,
|
| 29 |
+
"original_max_position_embeddings": 4096,
|
| 30 |
+
"pad_token_id": 199999,
|
| 31 |
+
"partial_rotary_factor": 0.75,
|
| 32 |
+
"quantization_config": {
|
| 33 |
+
"bits": 4,
|
| 34 |
+
"checkpoint_format": "gptq",
|
| 35 |
+
"desc_act": false,
|
| 36 |
+
"format": "gptq",
|
| 37 |
+
"group_size": 128,
|
| 38 |
+
"lm_head": false,
|
| 39 |
+
"meta": {
|
| 40 |
+
"act_group_aware": true,
|
| 41 |
+
"auto_forward_data_parallel": true,
|
| 42 |
+
"damp_auto_increment": 0.01,
|
| 43 |
+
"damp_percent": 0.05,
|
| 44 |
+
"dense_vram_strategy": "exclusive",
|
| 45 |
+
"dense_vram_strategy_devices": null,
|
| 46 |
+
"fallback": {
|
| 47 |
+
"smooth": null,
|
| 48 |
+
"strategy": "rtn",
|
| 49 |
+
"threshold": "0.5%"
|
| 50 |
+
},
|
| 51 |
+
"foem": null,
|
| 52 |
+
"gc_mode": "interval",
|
| 53 |
+
"gptaq": null,
|
| 54 |
+
"hessian": {
|
| 55 |
+
"chunk_bytes": null,
|
| 56 |
+
"chunk_size": null,
|
| 57 |
+
"staging_dtype": "float32"
|
| 58 |
+
},
|
| 59 |
+
"mock_quantization": false,
|
| 60 |
+
"moe_vram_strategy": "exclusive",
|
| 61 |
+
"moe_vram_strategy_devices": null,
|
| 62 |
+
"mse": 0.0,
|
| 63 |
+
"offload_to_disk": true,
|
| 64 |
+
"offload_to_disk_path": "/tmp/gptqmodel_l3mykc7c",
|
| 65 |
+
"pack_impl": "cpu",
|
| 66 |
+
"quantizer": [
|
| 67 |
+
"gptqmodel:7.0.0"
|
| 68 |
+
],
|
| 69 |
+
"static_groups": false,
|
| 70 |
+
"true_sequential": true,
|
| 71 |
+
"uri": "https://github.com/modelcloud/gptqmodel",
|
| 72 |
+
"wait_for_submodule_finalizers": false
|
| 73 |
+
},
|
| 74 |
+
"method": "gptq",
|
| 75 |
+
"pack_dtype": "int32",
|
| 76 |
+
"quant_method": "gptq",
|
| 77 |
+
"sym": true
|
| 78 |
+
},
|
| 79 |
+
"resid_pdrop": 0.0,
|
| 80 |
+
"rms_norm_eps": 1e-05,
|
| 81 |
+
"rope_parameters": {
|
| 82 |
+
"long_factor": [
|
| 83 |
+
1,
|
| 84 |
+
1.118320672,
|
| 85 |
+
1.250641126,
|
| 86 |
+
1.398617824,
|
| 87 |
+
1.564103225,
|
| 88 |
+
1.74916897,
|
| 89 |
+
1.956131817,
|
| 90 |
+
2.187582649,
|
| 91 |
+
2.446418898,
|
| 92 |
+
2.735880826,
|
| 93 |
+
3.059592084,
|
| 94 |
+
3.421605075,
|
| 95 |
+
3.826451687,
|
| 96 |
+
4.279200023,
|
| 97 |
+
4.785517845,
|
| 98 |
+
5.351743533,
|
| 99 |
+
5.984965424,
|
| 100 |
+
6.693110555,
|
| 101 |
+
7.485043894,
|
| 102 |
+
8.370679318,
|
| 103 |
+
9.36110372,
|
| 104 |
+
10.4687158,
|
| 105 |
+
11.70738129,
|
| 106 |
+
13.09260651,
|
| 107 |
+
14.64173252,
|
| 108 |
+
16.37415215,
|
| 109 |
+
18.31155283,
|
| 110 |
+
20.47818807,
|
| 111 |
+
22.90118105,
|
| 112 |
+
25.61086418,
|
| 113 |
+
28.64115884,
|
| 114 |
+
32.03,
|
| 115 |
+
32.1,
|
| 116 |
+
32.13,
|
| 117 |
+
32.23,
|
| 118 |
+
32.6,
|
| 119 |
+
32.61,
|
| 120 |
+
32.64,
|
| 121 |
+
32.66,
|
| 122 |
+
32.7,
|
| 123 |
+
32.71,
|
| 124 |
+
32.93,
|
| 125 |
+
32.97,
|
| 126 |
+
33.28,
|
| 127 |
+
33.49,
|
| 128 |
+
33.5,
|
| 129 |
+
44.16,
|
| 130 |
+
47.77
|
| 131 |
+
],
|
| 132 |
+
"original_max_position_embeddings": 4096,
|
| 133 |
+
"partial_rotary_factor": 0.75,
|
| 134 |
+
"rope_theta": 10000.0,
|
| 135 |
+
"rope_type": "longrope",
|
| 136 |
+
"short_factor": [
|
| 137 |
+
1.0,
|
| 138 |
+
1.0,
|
| 139 |
+
1.0,
|
| 140 |
+
1.0,
|
| 141 |
+
1.0,
|
| 142 |
+
1.0,
|
| 143 |
+
1.0,
|
| 144 |
+
1.0,
|
| 145 |
+
1.0,
|
| 146 |
+
1.0,
|
| 147 |
+
1.0,
|
| 148 |
+
1.0,
|
| 149 |
+
1.0,
|
| 150 |
+
1.0,
|
| 151 |
+
1.0,
|
| 152 |
+
1.0,
|
| 153 |
+
1.0,
|
| 154 |
+
1.0,
|
| 155 |
+
1.0,
|
| 156 |
+
1.0,
|
| 157 |
+
1.0,
|
| 158 |
+
1.0,
|
| 159 |
+
1.0,
|
| 160 |
+
1.0,
|
| 161 |
+
1.0,
|
| 162 |
+
1.0,
|
| 163 |
+
1.0,
|
| 164 |
+
1.0,
|
| 165 |
+
1.0,
|
| 166 |
+
1.0,
|
| 167 |
+
1.0,
|
| 168 |
+
1.0,
|
| 169 |
+
1.0,
|
| 170 |
+
1.0,
|
| 171 |
+
1.0,
|
| 172 |
+
1.0,
|
| 173 |
+
1.0,
|
| 174 |
+
1.0,
|
| 175 |
+
1.0,
|
| 176 |
+
1.0,
|
| 177 |
+
1.0,
|
| 178 |
+
1.0,
|
| 179 |
+
1.0,
|
| 180 |
+
1.0,
|
| 181 |
+
1.0,
|
| 182 |
+
1.0,
|
| 183 |
+
1.0,
|
| 184 |
+
1.0
|
| 185 |
+
],
|
| 186 |
+
"type": "longrope"
|
| 187 |
+
},
|
| 188 |
+
"sliding_window": 262144,
|
| 189 |
+
"tie_word_embeddings": true,
|
| 190 |
+
"transformers_version": "5.9.0",
|
| 191 |
+
"use_cache": true,
|
| 192 |
+
"vocab_size": 200064
|
| 193 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 199999,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
200020,
|
| 7 |
+
199999
|
| 8 |
+
],
|
| 9 |
+
"pad_token_id": 199999,
|
| 10 |
+
"transformers_version": "5.9.0"
|
| 11 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6e4959d65f52543658f0a1ccdd01ec423952432edb204c9f47b700b79bae89d2
|
| 3 |
+
size 2905820840
|
quant_log.csv
ADDED
|
@@ -0,0 +1,161 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
layer,module,loss,samples,damp,time
|
| 2 |
+
0,self_attn.qkv_proj,0.0000238235,0.05000,1.526
|
| 3 |
+
0,self_attn.o_proj,0.0000012382,0.05000,1.191
|
| 4 |
+
0,mlp.gate_proj,0.0000777876,0.05000,2.541
|
| 5 |
+
0,mlp.up_proj,0.0000749441,0.05000,2.631
|
| 6 |
+
0,mlp.down_proj,0.0000051147,0.05000,3.721
|
| 7 |
+
1,self_attn.qkv_proj,0.0001717266,0.05000,1.136
|
| 8 |
+
1,self_attn.o_proj,0.0000035207,0.05000,1.105
|
| 9 |
+
1,mlp.up_proj,0.0001190371,0.05000,2.573
|
| 10 |
+
1,mlp.gate_proj,0.0001262916,0.05000,2.590
|
| 11 |
+
1,mlp.down_proj,0.0000209985,0.05000,3.567
|
| 12 |
+
2,self_attn.qkv_proj,0.0004606717,0.05000,1.153
|
| 13 |
+
2,self_attn.o_proj,0.0000034317,0.05000,1.204
|
| 14 |
+
2,mlp.gate_proj,0.0001890496,0.05000,2.425
|
| 15 |
+
2,mlp.up_proj,0.0001766404,0.05000,2.478
|
| 16 |
+
2,mlp.down_proj,0.0000112479,0.05000,3.655
|
| 17 |
+
3,self_attn.qkv_proj,0.0006576798,0.05000,1.162
|
| 18 |
+
3,self_attn.o_proj,0.0000072551,0.05000,1.165
|
| 19 |
+
3,mlp.up_proj,0.0002476366,0.05000,2.459
|
| 20 |
+
3,mlp.gate_proj,0.0002709125,0.05000,2.520
|
| 21 |
+
3,mlp.down_proj,0.0259172664,0.05000,3.695
|
| 22 |
+
4,self_attn.qkv_proj,0.0006655008,0.05000,1.215
|
| 23 |
+
4,self_attn.o_proj,0.0000091820,0.05000,1.136
|
| 24 |
+
4,mlp.gate_proj,0.0003341130,0.05000,2.537
|
| 25 |
+
4,mlp.up_proj,0.0002933624,0.05000,2.573
|
| 26 |
+
4,mlp.down_proj,0.0000299716,0.05000,3.575
|
| 27 |
+
5,self_attn.qkv_proj,0.0007712968,0.05000,1.190
|
| 28 |
+
5,self_attn.o_proj,0.0000108736,0.05000,1.193
|
| 29 |
+
5,mlp.up_proj,0.0003416692,0.05000,2.371
|
| 30 |
+
5,mlp.gate_proj,0.0003962075,0.05000,2.424
|
| 31 |
+
5,mlp.down_proj,0.0000411710,0.05000,3.709
|
| 32 |
+
6,self_attn.qkv_proj,0.0006552734,0.05000,1.180
|
| 33 |
+
6,self_attn.o_proj,0.0000143339,0.05000,1.128
|
| 34 |
+
6,mlp.gate_proj,0.0003783428,0.05000,2.483
|
| 35 |
+
6,mlp.up_proj,0.0003409256,0.05000,2.500
|
| 36 |
+
6,mlp.down_proj,0.0000462377,0.05000,3.630
|
| 37 |
+
7,self_attn.qkv_proj,0.0009352742,0.05000,1.217
|
| 38 |
+
7,self_attn.o_proj,0.0000168276,0.05000,1.170
|
| 39 |
+
7,mlp.up_proj,0.0003627903,0.05000,2.417
|
| 40 |
+
7,mlp.gate_proj,0.0003876315,0.05000,2.451
|
| 41 |
+
7,mlp.down_proj,0.0000598990,0.05000,3.605
|
| 42 |
+
8,self_attn.qkv_proj,0.0008719424,0.05000,1.185
|
| 43 |
+
8,self_attn.o_proj,0.0000235102,0.05000,1.168
|
| 44 |
+
8,mlp.gate_proj,0.0003903917,0.05000,2.488
|
| 45 |
+
8,mlp.up_proj,0.0003862102,0.05000,2.522
|
| 46 |
+
8,mlp.down_proj,0.0000738566,0.05000,3.713
|
| 47 |
+
9,self_attn.qkv_proj,0.0008102300,0.05000,1.182
|
| 48 |
+
9,self_attn.o_proj,0.0000252474,0.05000,1.166
|
| 49 |
+
9,mlp.up_proj,0.0004095033,0.05000,2.477
|
| 50 |
+
9,mlp.gate_proj,0.0004039039,0.05000,2.545
|
| 51 |
+
9,mlp.down_proj,0.0000623042,0.05000,3.594
|
| 52 |
+
10,self_attn.qkv_proj,0.0008590046,0.05000,1.223
|
| 53 |
+
10,self_attn.o_proj,0.0000361045,0.05000,1.164
|
| 54 |
+
10,mlp.gate_proj,0.0004138994,0.05000,2.371
|
| 55 |
+
10,mlp.up_proj,0.0004185723,0.05000,2.445
|
| 56 |
+
10,mlp.down_proj,0.0000664324,0.05000,3.691
|
| 57 |
+
11,self_attn.qkv_proj,0.0008356616,0.05000,1.182
|
| 58 |
+
11,self_attn.o_proj,0.0000273909,0.05000,1.169
|
| 59 |
+
11,mlp.up_proj,0.0004351710,0.05000,2.454
|
| 60 |
+
11,mlp.gate_proj,0.0004140602,0.05000,2.499
|
| 61 |
+
11,mlp.down_proj,0.0000679036,0.05000,3.662
|
| 62 |
+
12,self_attn.qkv_proj,0.0008949989,0.05000,1.214
|
| 63 |
+
12,self_attn.o_proj,0.0000258759,0.05000,1.159
|
| 64 |
+
12,mlp.gate_proj,0.0003999382,0.05000,2.366
|
| 65 |
+
12,mlp.up_proj,0.0004371861,0.05000,2.444
|
| 66 |
+
12,mlp.down_proj,0.0000790061,0.05000,3.681
|
| 67 |
+
13,self_attn.qkv_proj,0.0010007589,0.05000,1.201
|
| 68 |
+
13,self_attn.o_proj,0.0000284145,0.05000,1.141
|
| 69 |
+
13,mlp.up_proj,0.0004744596,0.05000,2.411
|
| 70 |
+
13,mlp.gate_proj,0.0004218430,0.05000,2.477
|
| 71 |
+
13,mlp.down_proj,0.0000882345,0.05000,3.668
|
| 72 |
+
14,self_attn.qkv_proj,0.0009347553,0.05000,1.254
|
| 73 |
+
14,self_attn.o_proj,0.0000278657,0.05000,1.132
|
| 74 |
+
14,mlp.gate_proj,0.0004590117,0.05000,2.410
|
| 75 |
+
14,mlp.up_proj,0.0005133247,0.05000,2.452
|
| 76 |
+
14,mlp.down_proj,0.0000937629,0.05000,3.722
|
| 77 |
+
15,self_attn.qkv_proj,0.0010369427,0.05000,1.178
|
| 78 |
+
15,self_attn.o_proj,0.0000249127,0.05000,1.143
|
| 79 |
+
15,mlp.up_proj,0.0005674770,0.05000,2.445
|
| 80 |
+
15,mlp.gate_proj,0.0005078213,0.05000,2.458
|
| 81 |
+
15,mlp.down_proj,0.0001074678,0.05000,3.565
|
| 82 |
+
16,self_attn.qkv_proj,0.0009804815,0.05000,1.219
|
| 83 |
+
16,self_attn.o_proj,0.0000504799,0.05000,1.134
|
| 84 |
+
16,mlp.gate_proj,0.0005253795,0.05000,2.448
|
| 85 |
+
16,mlp.up_proj,0.0005916055,0.05000,2.500
|
| 86 |
+
16,mlp.down_proj,0.0001349178,0.05000,3.665
|
| 87 |
+
17,self_attn.qkv_proj,0.0010292538,0.05000,1.187
|
| 88 |
+
17,self_attn.o_proj,0.0000403605,0.05000,1.153
|
| 89 |
+
17,mlp.up_proj,0.0006312746,0.05000,2.306
|
| 90 |
+
17,mlp.gate_proj,0.0005545107,0.05000,2.378
|
| 91 |
+
17,mlp.down_proj,0.0001600333,0.05000,3.641
|
| 92 |
+
18,self_attn.qkv_proj,0.0010037017,0.05000,1.182
|
| 93 |
+
18,self_attn.o_proj,0.0000637205,0.05000,1.175
|
| 94 |
+
18,mlp.gate_proj,0.0005921678,0.05000,2.351
|
| 95 |
+
18,mlp.up_proj,0.0006664776,0.05000,2.437
|
| 96 |
+
18,mlp.down_proj,0.0002355822,0.05000,3.763
|
| 97 |
+
19,self_attn.qkv_proj,0.0010758908,0.05000,1.198
|
| 98 |
+
19,self_attn.o_proj,0.0000702092,0.05000,1.154
|
| 99 |
+
19,mlp.up_proj,0.0007063535,0.05000,2.483
|
| 100 |
+
19,mlp.gate_proj,0.0006349712,0.05000,2.577
|
| 101 |
+
19,mlp.down_proj,0.0002522909,0.05000,3.706
|
| 102 |
+
20,self_attn.qkv_proj,0.0010339192,0.05000,1.196
|
| 103 |
+
20,self_attn.o_proj,0.0000883794,0.05000,1.164
|
| 104 |
+
20,mlp.gate_proj,0.0007679065,0.05000,2.510
|
| 105 |
+
20,mlp.up_proj,0.0008546022,0.05000,2.524
|
| 106 |
+
20,mlp.down_proj,0.0003165363,0.05000,3.687
|
| 107 |
+
21,self_attn.qkv_proj,0.0012044958,0.05000,1.212
|
| 108 |
+
21,self_attn.o_proj,0.0000640508,0.05000,1.162
|
| 109 |
+
21,mlp.up_proj,0.0009123604,0.05000,2.478
|
| 110 |
+
21,mlp.gate_proj,0.0008116388,0.05000,2.527
|
| 111 |
+
21,mlp.down_proj,0.0004007476,0.05000,3.808
|
| 112 |
+
22,self_attn.qkv_proj,0.0011504276,0.05000,1.175
|
| 113 |
+
22,self_attn.o_proj,0.0000911345,0.05000,1.149
|
| 114 |
+
22,mlp.gate_proj,0.0008588976,0.05000,2.403
|
| 115 |
+
22,mlp.up_proj,0.0009739880,0.05000,2.458
|
| 116 |
+
22,mlp.down_proj,0.0004551394,0.05000,3.702
|
| 117 |
+
23,self_attn.qkv_proj,0.0013398652,0.05000,1.189
|
| 118 |
+
23,self_attn.o_proj,0.0000587729,0.05000,1.188
|
| 119 |
+
23,mlp.up_proj,0.0010498388,0.05000,2.402
|
| 120 |
+
23,mlp.gate_proj,0.0009337669,0.05000,2.449
|
| 121 |
+
23,mlp.down_proj,0.0004714720,0.05000,3.573
|
| 122 |
+
24,self_attn.qkv_proj,0.0014503187,0.05000,1.312
|
| 123 |
+
24,self_attn.o_proj,0.0000586708,0.05000,1.156
|
| 124 |
+
24,mlp.gate_proj,0.0010084476,0.05000,2.444
|
| 125 |
+
24,mlp.up_proj,0.0011423412,0.05000,2.492
|
| 126 |
+
24,mlp.down_proj,0.0005463777,0.05000,3.822
|
| 127 |
+
25,self_attn.qkv_proj,0.0015214244,0.05000,1.204
|
| 128 |
+
25,self_attn.o_proj,0.0000830141,0.05000,1.149
|
| 129 |
+
25,mlp.up_proj,0.0012727813,0.05000,2.471
|
| 130 |
+
25,mlp.gate_proj,0.0011201203,0.05000,2.545
|
| 131 |
+
25,mlp.down_proj,0.0006316668,0.05000,3.679
|
| 132 |
+
26,self_attn.qkv_proj,0.0013453441,0.05000,1.193
|
| 133 |
+
26,self_attn.o_proj,0.0001492357,0.05000,1.147
|
| 134 |
+
26,mlp.up_proj,0.0013456151,0.05000,2.563
|
| 135 |
+
26,mlp.gate_proj,0.0011857533,0.05000,2.593
|
| 136 |
+
26,mlp.down_proj,0.0007685589,0.05000,3.712
|
| 137 |
+
27,self_attn.qkv_proj,0.0014054913,0.05000,1.197
|
| 138 |
+
27,self_attn.o_proj,0.0001597724,0.05000,1.177
|
| 139 |
+
27,mlp.up_proj,0.0014349269,0.05000,2.501
|
| 140 |
+
27,mlp.gate_proj,0.0012792713,0.05000,2.572
|
| 141 |
+
27,mlp.down_proj,0.0009183092,0.05000,3.721
|
| 142 |
+
28,self_attn.qkv_proj,0.0016723366,0.05000,1.208
|
| 143 |
+
28,self_attn.o_proj,0.0001700563,0.05000,1.215
|
| 144 |
+
28,mlp.gate_proj,0.0013890771,0.05000,2.471
|
| 145 |
+
28,mlp.up_proj,0.0015182733,0.05000,2.542
|
| 146 |
+
28,mlp.down_proj,0.0011522665,0.05000,3.806
|
| 147 |
+
29,self_attn.qkv_proj,0.0015151091,0.05000,1.153
|
| 148 |
+
29,self_attn.o_proj,0.0002396083,0.05000,1.168
|
| 149 |
+
29,mlp.up_proj,0.0015197603,0.05000,2.468
|
| 150 |
+
29,mlp.gate_proj,0.0015041680,0.05000,2.508
|
| 151 |
+
29,mlp.down_proj,0.0018855077,0.05000,3.632
|
| 152 |
+
30,self_attn.qkv_proj,0.0010489146,0.05000,1.232
|
| 153 |
+
30,self_attn.o_proj,0.0002350578,0.05000,1.128
|
| 154 |
+
30,mlp.gate_proj,0.0015210857,0.05000,2.606
|
| 155 |
+
30,mlp.up_proj,0.0013783333,0.05000,2.638
|
| 156 |
+
30,mlp.down_proj,0.0171129781,0.05000,3.743
|
| 157 |
+
31,self_attn.qkv_proj,0.0000227647,0.05000,1.255
|
| 158 |
+
31,self_attn.o_proj,0.0000320150,0.05000,1.175
|
| 159 |
+
31,mlp.up_proj,0.0003993228,0.05000,2.532
|
| 160 |
+
31,mlp.gate_proj,0.0006483612,0.05000,2.584
|
| 161 |
+
31,mlp.down_proj,0.0030651457,0.05000,3.721
|
quantize_config.json
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bits": 4,
|
| 3 |
+
"group_size": 128,
|
| 4 |
+
"desc_act": false,
|
| 5 |
+
"lm_head": false,
|
| 6 |
+
"method": "gptq",
|
| 7 |
+
"quant_method": "gptq",
|
| 8 |
+
"format": "gptq",
|
| 9 |
+
"checkpoint_format": "gptq",
|
| 10 |
+
"pack_dtype": "int32",
|
| 11 |
+
"meta": {
|
| 12 |
+
"quantizer": [
|
| 13 |
+
"gptqmodel:7.0.0"
|
| 14 |
+
],
|
| 15 |
+
"uri": "https://github.com/modelcloud/gptqmodel",
|
| 16 |
+
"damp_percent": 0.05,
|
| 17 |
+
"damp_auto_increment": 0.01,
|
| 18 |
+
"static_groups": false,
|
| 19 |
+
"true_sequential": true,
|
| 20 |
+
"mse": 0.0,
|
| 21 |
+
"gptaq": null,
|
| 22 |
+
"foem": null,
|
| 23 |
+
"act_group_aware": true,
|
| 24 |
+
"fallback": {
|
| 25 |
+
"strategy": "rtn",
|
| 26 |
+
"threshold": "0.5%",
|
| 27 |
+
"smooth": null
|
| 28 |
+
},
|
| 29 |
+
"offload_to_disk": true,
|
| 30 |
+
"offload_to_disk_path": "/tmp/gptqmodel_l3mykc7c",
|
| 31 |
+
"pack_impl": "cpu",
|
| 32 |
+
"gc_mode": "interval",
|
| 33 |
+
"wait_for_submodule_finalizers": false,
|
| 34 |
+
"auto_forward_data_parallel": true,
|
| 35 |
+
"dense_vram_strategy": "exclusive",
|
| 36 |
+
"dense_vram_strategy_devices": null,
|
| 37 |
+
"moe_vram_strategy": "exclusive",
|
| 38 |
+
"moe_vram_strategy_devices": null,
|
| 39 |
+
"mock_quantization": false,
|
| 40 |
+
"hessian": {
|
| 41 |
+
"chunk_size": null,
|
| 42 |
+
"chunk_bytes": null,
|
| 43 |
+
"staging_dtype": "float32"
|
| 44 |
+
}
|
| 45 |
+
},
|
| 46 |
+
"sym": true
|
| 47 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:439894569b1646b4096ff6f0c104c2e078931c0581850932f8a576d05b755e5a
|
| 3 |
+
size 15524574
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|endoftext|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|endoftext|>",
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"model_max_length": 131072,
|
| 10 |
+
"pad_token": "<|endoftext|>",
|
| 11 |
+
"tokenizer_class": "TokenizersBackend",
|
| 12 |
+
"unk_token": "<|endoftext|>"
|
| 13 |
+
}
|