DopeorNope
commited on
Update config.json
Browse files- config.json +2 -16
config.json
CHANGED
@@ -6,37 +6,23 @@
|
|
6 |
"attention_bias": false,
|
7 |
"attention_dropout": 0.0,
|
8 |
"bos_token_id": 1,
|
9 |
-
"drop_attention_mask": false,
|
10 |
"eos_token_id": 2,
|
|
|
11 |
"hidden_act": "silu",
|
12 |
"hidden_size": 4096,
|
13 |
"initializer_range": 0.02,
|
14 |
"intermediate_size": 11008,
|
15 |
-
"mask_input": false,
|
16 |
-
"mask_rate": 0,
|
17 |
"max_position_embeddings": 4096,
|
18 |
-
"mixup": false,
|
19 |
"model_type": "llama",
|
20 |
-
"
|
21 |
-
"neftune_alpha": 0,
|
22 |
-
"not_mask_source": true,
|
23 |
-
"not_mask_tgt": false,
|
24 |
"num_attention_heads": 32,
|
25 |
"num_hidden_layers": 32,
|
26 |
"num_key_value_heads": 32,
|
27 |
-
"only_drop_target": false,
|
28 |
"pretraining_tp": 1,
|
29 |
-
"replace": false,
|
30 |
-
"replace_rate": 0,
|
31 |
"rms_norm_eps": 1e-05,
|
32 |
"rope_scaling": null,
|
33 |
"rope_theta": 10000.0,
|
34 |
"tie_word_embeddings": false,
|
35 |
-
"token_dropout": 0,
|
36 |
-
"token_noise": false,
|
37 |
-
"token_noise_hard": false,
|
38 |
-
"token_noise_random": false,
|
39 |
-
"token_noise_sample": false,
|
40 |
"torch_dtype": "float32",
|
41 |
"transformers_version": "4.36.2",
|
42 |
"use_cache": true,
|
|
|
6 |
"attention_bias": false,
|
7 |
"attention_dropout": 0.0,
|
8 |
"bos_token_id": 1,
|
|
|
9 |
"eos_token_id": 2,
|
10 |
+
"head_dim": 128,
|
11 |
"hidden_act": "silu",
|
12 |
"hidden_size": 4096,
|
13 |
"initializer_range": 0.02,
|
14 |
"intermediate_size": 11008,
|
|
|
|
|
15 |
"max_position_embeddings": 4096,
|
|
|
16 |
"model_type": "llama",
|
17 |
+
"mlp_bias": false,
|
|
|
|
|
|
|
18 |
"num_attention_heads": 32,
|
19 |
"num_hidden_layers": 32,
|
20 |
"num_key_value_heads": 32,
|
|
|
21 |
"pretraining_tp": 1,
|
|
|
|
|
22 |
"rms_norm_eps": 1e-05,
|
23 |
"rope_scaling": null,
|
24 |
"rope_theta": 10000.0,
|
25 |
"tie_word_embeddings": false,
|
|
|
|
|
|
|
|
|
|
|
26 |
"torch_dtype": "float32",
|
27 |
"transformers_version": "4.36.2",
|
28 |
"use_cache": true,
|