Update config.json
Browse files- config.json +57 -57
config.json
CHANGED
@@ -1,57 +1,57 @@
|
|
1 |
-
{
|
2 |
-
"activation_function": "swiglu",
|
3 |
-
"architectures": [
|
4 |
-
"NomicBertModel"
|
5 |
-
],
|
6 |
-
"attn_pdrop": 0.0,
|
7 |
-
"auto_map": {
|
8 |
-
"AutoConfig": "
|
9 |
-
"AutoModel": "
|
10 |
-
"AutoModelForMaskedLM": "
|
11 |
-
},
|
12 |
-
"bos_token_id": null,
|
13 |
-
"causal": false,
|
14 |
-
"dense_seq_output": true,
|
15 |
-
"embd_pdrop": 0.0,
|
16 |
-
"eos_token_id": null,
|
17 |
-
"fused_bias_fc": true,
|
18 |
-
"fused_dropout_add_ln": true,
|
19 |
-
"initializer_range": 0.02,
|
20 |
-
"layer_norm_epsilon": 1e-12,
|
21 |
-
"max_trained_positions": 2048,
|
22 |
-
"mlp_fc1_bias": false,
|
23 |
-
"mlp_fc2_bias": false,
|
24 |
-
"model_type": "nomic_bert",
|
25 |
-
"n_embd": 768,
|
26 |
-
"n_head": 12,
|
27 |
-
"n_inner": 3072,
|
28 |
-
"n_layer": 12,
|
29 |
-
"n_positions": 8192,
|
30 |
-
"pad_vocab_size_multiple": 64,
|
31 |
-
"parallel_block": false,
|
32 |
-
"parallel_block_tied_norm": false,
|
33 |
-
"prenorm": false,
|
34 |
-
"qkv_proj_bias": false,
|
35 |
-
"reorder_and_upcast_attn": false,
|
36 |
-
"resid_pdrop": 0.0,
|
37 |
-
"rotary_emb_base": 1000,
|
38 |
-
"rotary_emb_fraction": 1.0,
|
39 |
-
"rotary_emb_interleaved": false,
|
40 |
-
"rotary_emb_scale_base": null,
|
41 |
-
"rotary_scaling_factor": null,
|
42 |
-
"scale_attn_by_inverse_layer_idx": false,
|
43 |
-
"scale_attn_weights": true,
|
44 |
-
"summary_activation": null,
|
45 |
-
"summary_first_dropout": 0.0,
|
46 |
-
"summary_proj_to_labels": true,
|
47 |
-
"summary_type": "cls_index",
|
48 |
-
"summary_use_proj": true,
|
49 |
-
"torch_dtype": "float32",
|
50 |
-
"transformers_version": "4.37.2",
|
51 |
-
"type_vocab_size": 2,
|
52 |
-
"use_cache": true,
|
53 |
-
"use_flash_attn": true,
|
54 |
-
"use_rms_norm": false,
|
55 |
-
"use_xentropy": true,
|
56 |
-
"vocab_size": 30528
|
57 |
-
}
|
|
|
1 |
+
{
|
2 |
+
"activation_function": "swiglu",
|
3 |
+
"architectures": [
|
4 |
+
"NomicBertModel"
|
5 |
+
],
|
6 |
+
"attn_pdrop": 0.0,
|
7 |
+
"auto_map": {
|
8 |
+
"AutoConfig": "BASF-AI/nomic-bert-2048--configuration_hf_nomic_bert.NomicBertConfig",
|
9 |
+
"AutoModel": "BASF-AI/nomic-bert-2048--modeling_hf_nomic_bert.NomicBertModel",
|
10 |
+
"AutoModelForMaskedLM": "BASF-AI/nomic-bert-2048--modeling_hf_nomic_bert.NomicBertForPreTraining"
|
11 |
+
},
|
12 |
+
"bos_token_id": null,
|
13 |
+
"causal": false,
|
14 |
+
"dense_seq_output": true,
|
15 |
+
"embd_pdrop": 0.0,
|
16 |
+
"eos_token_id": null,
|
17 |
+
"fused_bias_fc": true,
|
18 |
+
"fused_dropout_add_ln": true,
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"layer_norm_epsilon": 1e-12,
|
21 |
+
"max_trained_positions": 2048,
|
22 |
+
"mlp_fc1_bias": false,
|
23 |
+
"mlp_fc2_bias": false,
|
24 |
+
"model_type": "nomic_bert",
|
25 |
+
"n_embd": 768,
|
26 |
+
"n_head": 12,
|
27 |
+
"n_inner": 3072,
|
28 |
+
"n_layer": 12,
|
29 |
+
"n_positions": 8192,
|
30 |
+
"pad_vocab_size_multiple": 64,
|
31 |
+
"parallel_block": false,
|
32 |
+
"parallel_block_tied_norm": false,
|
33 |
+
"prenorm": false,
|
34 |
+
"qkv_proj_bias": false,
|
35 |
+
"reorder_and_upcast_attn": false,
|
36 |
+
"resid_pdrop": 0.0,
|
37 |
+
"rotary_emb_base": 1000,
|
38 |
+
"rotary_emb_fraction": 1.0,
|
39 |
+
"rotary_emb_interleaved": false,
|
40 |
+
"rotary_emb_scale_base": null,
|
41 |
+
"rotary_scaling_factor": null,
|
42 |
+
"scale_attn_by_inverse_layer_idx": false,
|
43 |
+
"scale_attn_weights": true,
|
44 |
+
"summary_activation": null,
|
45 |
+
"summary_first_dropout": 0.0,
|
46 |
+
"summary_proj_to_labels": true,
|
47 |
+
"summary_type": "cls_index",
|
48 |
+
"summary_use_proj": true,
|
49 |
+
"torch_dtype": "float32",
|
50 |
+
"transformers_version": "4.37.2",
|
51 |
+
"type_vocab_size": 2,
|
52 |
+
"use_cache": true,
|
53 |
+
"use_flash_attn": true,
|
54 |
+
"use_rms_norm": false,
|
55 |
+
"use_xentropy": true,
|
56 |
+
"vocab_size": 30528
|
57 |
+
}
|