zhijunjunlin commited on
Commit
b93388e
·
verified ·
1 Parent(s): 8a0b9ad

Training in progress, step 500

Browse files
config.json CHANGED
@@ -1,27 +1,46 @@
1
  {
2
- "_name_or_path": "distilroberta-base",
 
 
3
  "architectures": [
4
- "RobertaForMaskedLM"
5
  ],
6
- "attention_probs_dropout_prob": 0.1,
7
- "bos_token_id": 0,
8
- "classifier_dropout": null,
9
- "eos_token_id": 2,
10
- "hidden_act": "gelu",
11
- "hidden_dropout_prob": 0.1,
12
- "hidden_size": 768,
13
  "initializer_range": 0.02,
14
- "intermediate_size": 3072,
15
- "layer_norm_eps": 1e-05,
16
- "max_position_embeddings": 514,
17
- "model_type": "roberta",
18
- "num_attention_heads": 12,
19
- "num_hidden_layers": 6,
20
- "pad_token_id": 1,
21
- "position_embedding_type": "absolute",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
22
  "torch_dtype": "float32",
23
  "transformers_version": "4.40.0.dev0",
24
- "type_vocab_size": 1,
25
  "use_cache": true,
26
- "vocab_size": 50265
27
  }
 
1
  {
2
+ "_name_or_path": "distilgpt2",
3
+ "_num_labels": 1,
4
+ "activation_function": "gelu_new",
5
  "architectures": [
6
+ "GPT2LMHeadModel"
7
  ],
8
+ "attn_pdrop": 0.1,
9
+ "bos_token_id": 50256,
10
+ "embd_pdrop": 0.1,
11
+ "eos_token_id": 50256,
12
+ "id2label": {
13
+ "0": "LABEL_0"
14
+ },
15
  "initializer_range": 0.02,
16
+ "label2id": {
17
+ "LABEL_0": 0
18
+ },
19
+ "layer_norm_epsilon": 1e-05,
20
+ "model_type": "gpt2",
21
+ "n_ctx": 1024,
22
+ "n_embd": 768,
23
+ "n_head": 12,
24
+ "n_inner": null,
25
+ "n_layer": 6,
26
+ "n_positions": 1024,
27
+ "reorder_and_upcast_attn": false,
28
+ "resid_pdrop": 0.1,
29
+ "scale_attn_by_inverse_layer_idx": false,
30
+ "scale_attn_weights": true,
31
+ "summary_activation": null,
32
+ "summary_first_dropout": 0.1,
33
+ "summary_proj_to_labels": true,
34
+ "summary_type": "cls_index",
35
+ "summary_use_proj": true,
36
+ "task_specific_params": {
37
+ "text-generation": {
38
+ "do_sample": true,
39
+ "max_length": 50
40
+ }
41
+ },
42
  "torch_dtype": "float32",
43
  "transformers_version": "4.40.0.dev0",
 
44
  "use_cache": true,
45
+ "vocab_size": 50257
46
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4727c0565afb477fd17f1fd8fda7942d14e5d0ce7883559d7628d6405305d74a
3
- size 328693404
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99b751e2928afb4efa2c46269cac04e54ad09fa15866675a8253600be6d8ac42
3
+ size 327657928
runs/Apr05_02-05-27_DESKTOP-RHJ1ESI/events.out.tfevents.1712253943.DESKTOP-RHJ1ESI.208972.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:464272a301ecae956a3ec77421ca9bee25fba801b737e48830c1fcb3b5ec5146
3
- size 8874
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cedd505e97270f6d4a798def34d32171733659eee14c359dc3adff81eaa5e2d
3
+ size 9145
runs/Apr07_05-27-25_DESKTOP-RHJ1ESI/events.out.tfevents.1712438854.DESKTOP-RHJ1ESI.222551.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b4050a1e3d425eaac835228a13296f6a9aed97b4824c1b2f28393d8adaccd0c
3
+ size 5076
runs/Apr07_05-33-21_DESKTOP-RHJ1ESI/events.out.tfevents.1712439211.DESKTOP-RHJ1ESI.222551.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:976b010be91df19eb10acc3e517027f975581907257c0837520517e78ad8fbdf
3
+ size 5076
runs/Apr07_05-49-56_DESKTOP-RHJ1ESI/events.out.tfevents.1712440198.DESKTOP-RHJ1ESI.222551.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4247013abf177c9e51de553603fa7af2f8c50f1f72e949b9e7339cf8ac1acd0a
3
+ size 5076
runs/Apr07_05-59-27_DESKTOP-RHJ1ESI/events.out.tfevents.1712440768.DESKTOP-RHJ1ESI.222551.3 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3adad700e58db0c957134d13f1b55ece2d19c44966ef80f799f3708a5cd39392
3
+ size 5076
runs/Apr07_06-03-56_DESKTOP-RHJ1ESI/events.out.tfevents.1712441040.DESKTOP-RHJ1ESI.222551.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2d23f5cfd636a8051b36a3c6b945797c0f076243b81156e114249a7e8626c93
3
+ size 5076
runs/Apr07_06-07-13_DESKTOP-RHJ1ESI/events.out.tfevents.1712441234.DESKTOP-RHJ1ESI.222551.5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c29c18215f9ebf1af2288825672a820b627e49b271b47871a1d46a9719358dc
3
+ size 5076
runs/Apr07_17-23-26_DESKTOP-RHJ1ESI/events.out.tfevents.1712481813.DESKTOP-RHJ1ESI.222551.7 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71cabd41fd5dbbab2ab48735ae464522a05883e388ac96ec21d317fc1119b8d2
3
+ size 5076
runs/Apr07_17-39-01_DESKTOP-RHJ1ESI/events.out.tfevents.1712482742.DESKTOP-RHJ1ESI.233984.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3783aa4800c998c20a4194ae09ec9962fe6265b2b8a6edb71c52b1fcaec9e54d
3
+ size 5076
runs/Apr07_17-55-38_DESKTOP-RHJ1ESI/events.out.tfevents.1712483741.DESKTOP-RHJ1ESI.234351.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e4bb0e28bb8d80d84858b02307741a5ee0213eab00b00dde4fc59cc67eec9bc
3
+ size 5287
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c8882b493a4d0b536f2acd4c3773536ec7d328d0c4119c650c8a0ad71faccd17
3
  size 5048
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9192b4808ac2c0f55eb20bc280e2ce7fa7f49d6438459329c8818a7c97fcb1a7
3
  size 5048