Update config.json
Browse files- config.json +20 -0
config.json
CHANGED
|
@@ -3,6 +3,26 @@
|
|
| 3 |
"architectures": [
|
| 4 |
"Qwen2ForCausalLM"
|
| 5 |
],
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
"attention_dropout": 0.0,
|
| 7 |
"bos_token_id": 151643,
|
| 8 |
"eos_token_id": 151645,
|
|
|
|
| 3 |
"architectures": [
|
| 4 |
"Qwen2ForCausalLM"
|
| 5 |
],
|
| 6 |
+
"quantization_config": {
|
| 7 |
+
"bits": 8,
|
| 8 |
+
"dynamic": null,
|
| 9 |
+
"group_size": 128,
|
| 10 |
+
"desc_act": true,
|
| 11 |
+
"sym": true,
|
| 12 |
+
"lm_head": false,
|
| 13 |
+
"quant_method": "gptq",
|
| 14 |
+
"checkpoint_format": "gptq",
|
| 15 |
+
"meta": {
|
| 16 |
+
"quantizer": [
|
| 17 |
+
"gptqmodel:1.7.4"
|
| 18 |
+
],
|
| 19 |
+
"uri": "https://github.com/modelcloud/gptqmodel",
|
| 20 |
+
"damp_percent": 0.01,
|
| 21 |
+
"damp_auto_increment": 0.0025,
|
| 22 |
+
"static_groups": false,
|
| 23 |
+
"true_sequential": true
|
| 24 |
+
}
|
| 25 |
+
},
|
| 26 |
"attention_dropout": 0.0,
|
| 27 |
"bos_token_id": 151643,
|
| 28 |
"eos_token_id": 151645,
|