CyrilMa commited on
Commit
7ed4994
·
verified ·
1 Parent(s): 1036956

Upload folder using huggingface_hub

Browse files
config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "gelu",
3
+ "architectures": [
4
+ "ProteomeLMForMaskedLM"
5
+ ],
6
+ "attention_dropout": 0.1,
7
+ "batch_size": 64,
8
+ "beta1": 0.9,
9
+ "beta2": 0.999,
10
+ "dataloader_num_workers": 8,
11
+ "db_path": "/home/malbrank/training/",
12
+ "dim": 768,
13
+ "dropout": 0.1,
14
+ "dtype": "bfloat16",
15
+ "eval_accumulation_steps": 1,
16
+ "eval_epochs": 1,
17
+ "gradient_accumulation_steps": 1,
18
+ "gradient_checkpointing": false,
19
+ "hidden_dim": 768,
20
+ "initializer_range": 0.02,
21
+ "input_size": 1152,
22
+ "learning_rate": 0.0005,
23
+ "logging_steps": 10,
24
+ "loss_type": "mse",
25
+ "mask_fraction": 0.5,
26
+ "max_grad_norm": 1.0,
27
+ "max_position_embeddings": 512,
28
+ "max_steps": 1000000,
29
+ "min_taxid_size": 200,
30
+ "model_type": "distilbert",
31
+ "n_heads": 12,
32
+ "n_inner": null,
33
+ "n_layers": 12,
34
+ "namedir": "ProteomeLM-S-Kuma",
35
+ "num_epochs": 2000,
36
+ "output_dir": "/home/malbrank/ProteomeLM/output/",
37
+ "pad_token_id": 0,
38
+ "qa_dropout": 0.1,
39
+ "save_epochs": 15,
40
+ "scheduler": "cosine",
41
+ "seq_classif_dropout": 0.2,
42
+ "sinusoidal_pos_embds": false,
43
+ "torch_dtype": "bfloat16",
44
+ "transformers_version": "4.51.1",
45
+ "use_cache": true,
46
+ "use_one_gpu": "0",
47
+ "vocab_size": 30522,
48
+ "wandb_project": "proteomelm",
49
+ "warmup_steps": 500,
50
+ "weight_decay": 0.01
51
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2dfd6ebd8a0cd054e199257f32882af996c89e1bb31cb736ce16ac9da4a8407f
3
+ size 224516316
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65798619ac450a2b54834bee9504588718836b2dafa49c568977e6228b498d80
3
+ size 181030465
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:384f92eb646333a5eaa765f22f3a0642ded89ecd015fd861c0288a007ba3deb5
3
+ size 14443
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98f11a9bc434491ee8f2ee7a14b652a2c298a57b1f45cfb70ce103c73207bb68
3
+ size 1263
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee6b400d735ea213553755932ee9dd2611ece0fdd117e80505b959b52e5f2d08
3
+ size 5571