Upload constant weights as EsmForMaskedLM
Browse files- config.json +28 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "./6_large-scale/mixed/models/mxd-const-0.625_<sep>_lr1e-4_650M-ESM_500k-stp_2025-06-09/",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"EsmForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"emb_layer_norm_before": null,
|
| 8 |
+
"esmfold_config": null,
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 1280,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 5120,
|
| 13 |
+
"is_folding_model": false,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"mask_token_id": 32,
|
| 16 |
+
"max_position_embeddings": 322,
|
| 17 |
+
"model_type": "esm",
|
| 18 |
+
"num_attention_heads": 20,
|
| 19 |
+
"num_hidden_layers": 33,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "rotary",
|
| 22 |
+
"token_dropout": false,
|
| 23 |
+
"torch_dtype": "float32",
|
| 24 |
+
"transformers_version": "4.44.2",
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_list": null,
|
| 27 |
+
"vocab_size": 33
|
| 28 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:245ac8de0c7f0adb80302178662e016abb601d252d3b0162ce155af1b8835ce9
|
| 3 |
+
size 2605893600
|