danielhanchen
commited on
Commit
•
2774aa7
1
Parent(s):
bde0d21
Upload config
Browse files- config.json +1 -16
config.json
CHANGED
@@ -18,21 +18,6 @@
|
|
18 |
"num_hidden_layers": 80,
|
19 |
"num_key_value_heads": 8,
|
20 |
"pretraining_tp": 1,
|
21 |
-
"quantization_config": {
|
22 |
-
"_load_in_4bit": true,
|
23 |
-
"_load_in_8bit": false,
|
24 |
-
"bnb_4bit_compute_dtype": "bfloat16",
|
25 |
-
"bnb_4bit_quant_storage": "uint8",
|
26 |
-
"bnb_4bit_quant_type": "nf4",
|
27 |
-
"bnb_4bit_use_double_quant": true,
|
28 |
-
"llm_int8_enable_fp32_cpu_offload": false,
|
29 |
-
"llm_int8_has_fp16_weight": false,
|
30 |
-
"llm_int8_skip_modules": null,
|
31 |
-
"llm_int8_threshold": 6.0,
|
32 |
-
"load_in_4bit": true,
|
33 |
-
"load_in_8bit": false,
|
34 |
-
"quant_method": "bitsandbytes"
|
35 |
-
},
|
36 |
"rms_norm_eps": 1e-05,
|
37 |
"rope_scaling": {
|
38 |
"factor": 8.0,
|
@@ -44,7 +29,7 @@
|
|
44 |
"rope_theta": 500000.0,
|
45 |
"tie_word_embeddings": false,
|
46 |
"torch_dtype": "bfloat16",
|
47 |
-
"transformers_version": "4.
|
48 |
"use_cache": true,
|
49 |
"vocab_size": 128256
|
50 |
}
|
|
|
18 |
"num_hidden_layers": 80,
|
19 |
"num_key_value_heads": 8,
|
20 |
"pretraining_tp": 1,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
21 |
"rms_norm_eps": 1e-05,
|
22 |
"rope_scaling": {
|
23 |
"factor": 8.0,
|
|
|
29 |
"rope_theta": 500000.0,
|
30 |
"tie_word_embeddings": false,
|
31 |
"torch_dtype": "bfloat16",
|
32 |
+
"transformers_version": "4.44.0",
|
33 |
"use_cache": true,
|
34 |
"vocab_size": 128256
|
35 |
}
|