Upload config.json with huggingface_hub
Browse files- config.json +2 -3
config.json
CHANGED
|
@@ -26,7 +26,6 @@
|
|
| 26 |
"lm_head"
|
| 27 |
],
|
| 28 |
"export": {
|
| 29 |
-
"custom_mode": "fp8",
|
| 30 |
"kv_cache_group": [],
|
| 31 |
"pack_method": "reorder",
|
| 32 |
"weight_format": "real_quantized",
|
|
@@ -73,14 +72,14 @@
|
|
| 73 |
"layer_type_quant_config": {},
|
| 74 |
"pack_method": "reorder",
|
| 75 |
"quant_method": "quark",
|
| 76 |
-
"quant_mode":
|
| 77 |
},
|
| 78 |
"rms_norm_eps": 1e-06,
|
| 79 |
"rope_scaling": null,
|
| 80 |
"rope_theta": 10000.0,
|
| 81 |
"tie_word_embeddings": false,
|
| 82 |
"torch_dtype": "float16",
|
| 83 |
-
"transformers_version": "4.
|
| 84 |
"use_cache": true,
|
| 85 |
"vocab_size": 32000
|
| 86 |
}
|
|
|
|
| 26 |
"lm_head"
|
| 27 |
],
|
| 28 |
"export": {
|
|
|
|
| 29 |
"kv_cache_group": [],
|
| 30 |
"pack_method": "reorder",
|
| 31 |
"weight_format": "real_quantized",
|
|
|
|
| 72 |
"layer_type_quant_config": {},
|
| 73 |
"pack_method": "reorder",
|
| 74 |
"quant_method": "quark",
|
| 75 |
+
"quant_mode": "eager_mode"
|
| 76 |
},
|
| 77 |
"rms_norm_eps": 1e-06,
|
| 78 |
"rope_scaling": null,
|
| 79 |
"rope_theta": 10000.0,
|
| 80 |
"tie_word_embeddings": false,
|
| 81 |
"torch_dtype": "float16",
|
| 82 |
+
"transformers_version": "4.45.2",
|
| 83 |
"use_cache": true,
|
| 84 |
"vocab_size": 32000
|
| 85 |
}
|