Upload WhisperForConditionalGeneration
Browse files- config.json +16 -1
- model.safetensors +3 -0
config.json
CHANGED
|
@@ -42,8 +42,23 @@
|
|
| 42 |
"num_hidden_layers": 32,
|
| 43 |
"num_mel_bins": 80,
|
| 44 |
"pad_token_id": 50257,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
"scale_embedding": false,
|
| 46 |
-
"torch_dtype": "
|
| 47 |
"transformers_version": "4.42.0.dev0",
|
| 48 |
"use_cache": false,
|
| 49 |
"use_weighted_layer_sum": false,
|
|
|
|
| 42 |
"num_hidden_layers": 32,
|
| 43 |
"num_mel_bins": 80,
|
| 44 |
"pad_token_id": 50257,
|
| 45 |
+
"quantization_config": {
|
| 46 |
+
"_load_in_4bit": false,
|
| 47 |
+
"_load_in_8bit": true,
|
| 48 |
+
"bnb_4bit_compute_dtype": "float32",
|
| 49 |
+
"bnb_4bit_quant_storage": "uint8",
|
| 50 |
+
"bnb_4bit_quant_type": "fp4",
|
| 51 |
+
"bnb_4bit_use_double_quant": false,
|
| 52 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
| 53 |
+
"llm_int8_has_fp16_weight": false,
|
| 54 |
+
"llm_int8_skip_modules": null,
|
| 55 |
+
"llm_int8_threshold": 6.0,
|
| 56 |
+
"load_in_4bit": false,
|
| 57 |
+
"load_in_8bit": true,
|
| 58 |
+
"quant_method": "bitsandbytes"
|
| 59 |
+
},
|
| 60 |
"scale_embedding": false,
|
| 61 |
+
"torch_dtype": "float16",
|
| 62 |
"transformers_version": "4.42.0.dev0",
|
| 63 |
"use_cache": false,
|
| 64 |
"use_weighted_layer_sum": false,
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2bc1e70acae0986367c6cd4292c8044090c319f904b234d758755172c8e30e1
|
| 3 |
+
size 1622472984
|