negfir commited on
Commit
c1d8c05
1 Parent(s): aa910cf

Training in progress, step 500

Browse files
.gitignore ADDED
@@ -0,0 +1 @@
 
 
1
+ checkpoint-*/
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "squeezebert/squeezebert-uncased",
3
+ "architectures": [
4
+ "SqueezeBertForMaskedLM"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "embedding_size": 768,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "initializer_range": 0.02,
12
+ "intermediate_groups": 4,
13
+ "intermediate_size": 3072,
14
+ "k_groups": 4,
15
+ "layer_norm_eps": 1e-12,
16
+ "max_position_embeddings": 512,
17
+ "model_type": "squeezebert",
18
+ "num_attention_heads": 12,
19
+ "num_hidden_layers": 10,
20
+ "output_groups": 4,
21
+ "pad_token_id": 0,
22
+ "post_attention_groups": 1,
23
+ "q_groups": 4,
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.17.0",
26
+ "type_vocab_size": 2,
27
+ "v_groups": 4,
28
+ "vocab_size": 30528
29
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0954cd8a1040e146831e5c0815b1bcf098c2b9bee772cb29ab9ecf2894716b7
3
+ size 189170407
runs/Mar08_17-01-41_fe18a385b66e/1646758926.5946379/events.out.tfevents.1646758926.fe18a385b66e.80.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6dd5f7de3c947a07537f73c78bdbab8594f14cf8a0fecb333f7bb131cc4efd0
3
+ size 4758
runs/Mar08_17-01-41_fe18a385b66e/events.out.tfevents.1646758926.fe18a385b66e.80.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a82f49d3a98048764f9db635fe1b5ac60355969ba8dde026f51928e0498dde4
3
+ size 3455
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:497115be3850f40e353e0524a11867b43d93208d5907118e1bf1cf61e11c9bb7
3
+ size 3055