WarrenWang01 commited on
Commit
bf56f97
·
verified ·
1 Parent(s): c1a93ad

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +31 -26
README.md CHANGED
@@ -1,41 +1,46 @@
1
-
 
 
 
 
 
 
 
 
2
 
3
  ## Training and evaluation data
4
  ### Training data
5
- NaturalQuestions,TriviaQA,WebQuestions
 
 
6
 
7
  ### Evaluation data
8
- NaturalQuestions,TriviaQA,WebQuestions,HotpotQA,PopQA
 
 
 
 
9
 
10
 
11
  ## Training hyperparameters
12
  The following hyperparameters were used during training:
13
  ### SFT
14
- learning_rate: 1e-6
15
- train_batch_size: 256
16
- train_micro_batch_size: 16
17
- optimizer: Use adamw_torch with betas=(0.9,0.95)
18
- lr_scheduler: cosine
19
- num_epochs: 8.0
20
 
21
  ## RL
22
- algorithm: dapo
23
- learning_rate: 1e-7
24
- lr_warmup_steps: 10
25
- num_epochs: 1.0
26
 
27
  ## Framework versions
28
- Transformers 4.57.6
29
- Pytorch 2.4.0
30
- Vllm 0.6.3
31
- flash-attn 2.7.4.post1
32
 
33
- ---
34
- license: apache-2.0
35
- metrics:
36
- - exact_match
37
- - f1
38
- - rouge
39
- tags:
40
- - GRIP
41
- ---
 
1
+ ---
2
+ license: apache-2.0
3
+ metrics:
4
+ - exact_match
5
+ - f1
6
+ - rouge
7
+ tags:
8
+ - GRIP
9
+ ---
10
 
11
  ## Training and evaluation data
12
  ### Training data
13
+ - NaturalQuestions
14
+ - TriviaQA
15
+ - WebQuestions
16
 
17
  ### Evaluation data
18
+ - NaturalQuestions
19
+ - TriviaQA
20
+ - WebQuestions
21
+ - HotpotQA
22
+ - PopQA
23
 
24
 
25
  ## Training hyperparameters
26
  The following hyperparameters were used during training:
27
  ### SFT
28
+ - learning_rate: 1e-6
29
+ - train_batch_size: 256
30
+ - train_micro_batch_size: 16
31
+ - optimizer: Use adamw_torch with betas=(0.9,0.95)
32
+ - lr_scheduler: cosine
33
+ - num_epochs: 8.0
34
 
35
  ## RL
36
+ - algorithm: dapo
37
+ - learning_rate: 1e-7
38
+ - lr_warmup_steps: 10
39
+ - num_epochs: 1.0
40
 
41
  ## Framework versions
42
+ - Transformers 4.57.6
43
+ - Pytorch 2.4.0
44
+ - Vllm 0.6.3
45
+ - flash-attn 2.7.4.post1
46