Update transformer/config.json
Browse files- transformer/config.json +16 -1
transformer/config.json
CHANGED
|
@@ -16,5 +16,20 @@
|
|
| 16 |
"num_single_layers": 38,
|
| 17 |
"out_channels": null,
|
| 18 |
"patch_size": 1,
|
| 19 |
-
"pooled_projection_dim": 768
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
}
|
|
|
|
| 16 |
"num_single_layers": 38,
|
| 17 |
"out_channels": null,
|
| 18 |
"patch_size": 1,
|
| 19 |
+
"pooled_projection_dim": 768,
|
| 20 |
+
"quantization_config": {
|
| 21 |
+
"_load_in_4bit": true,
|
| 22 |
+
"_load_in_8bit": false,
|
| 23 |
+
"bnb_4bit_compute_dtype": "bfloat16",
|
| 24 |
+
"bnb_4bit_quant_storage": "uint8",
|
| 25 |
+
"bnb_4bit_quant_type": "nf4",
|
| 26 |
+
"bnb_4bit_use_double_quant": false,
|
| 27 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
| 28 |
+
"llm_int8_has_fp16_weight": false,
|
| 29 |
+
"llm_int8_skip_modules": null,
|
| 30 |
+
"llm_int8_threshold": 6.0,
|
| 31 |
+
"load_in_4bit": true,
|
| 32 |
+
"load_in_8bit": false,
|
| 33 |
+
"quant_method": "bitsandbytes"
|
| 34 |
+
}
|
| 35 |
}
|