Upload llm/config.json with huggingface_hub
Browse files- llm/config.json +0 -15
llm/config.json
CHANGED
|
@@ -21,21 +21,6 @@
|
|
| 21 |
"num_key_value_heads": 20,
|
| 22 |
"pad_token_id": 0,
|
| 23 |
"pretraining_tp": 1,
|
| 24 |
-
"quantization_config": {
|
| 25 |
-
"_load_in_4bit": true,
|
| 26 |
-
"_load_in_8bit": false,
|
| 27 |
-
"bnb_4bit_compute_dtype": "float16",
|
| 28 |
-
"bnb_4bit_quant_storage": "uint8",
|
| 29 |
-
"bnb_4bit_quant_type": "nf4",
|
| 30 |
-
"bnb_4bit_use_double_quant": true,
|
| 31 |
-
"llm_int8_enable_fp32_cpu_offload": false,
|
| 32 |
-
"llm_int8_has_fp16_weight": false,
|
| 33 |
-
"llm_int8_skip_modules": null,
|
| 34 |
-
"llm_int8_threshold": 6.0,
|
| 35 |
-
"load_in_4bit": true,
|
| 36 |
-
"load_in_8bit": false,
|
| 37 |
-
"quant_method": "bitsandbytes"
|
| 38 |
-
},
|
| 39 |
"rms_norm_eps": 1e-05,
|
| 40 |
"rope_scaling": null,
|
| 41 |
"rope_theta": 10000.0,
|
|
|
|
| 21 |
"num_key_value_heads": 20,
|
| 22 |
"pad_token_id": 0,
|
| 23 |
"pretraining_tp": 1,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
"rms_norm_eps": 1e-05,
|
| 25 |
"rope_scaling": null,
|
| 26 |
"rope_theta": 10000.0,
|