Hugging Face's logo

cy948
/
starcoder-peft-airscript

PEFT
Safetensors
Generated from Trainer
Model card Files Community
cy948 commited on
Commit
1b6a221
·
verified ·
1 Parent(s): 53f126d

End of training

Browse files
Files changed (1) hide show
  1. README.md +18 -16
README.md CHANGED
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [bigcode/starcoderbase-1b](https://huggingface.co/bigcode/starcoderbase-1b) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 0.8171
20
 
21
  ## Model description
22
 
@@ -44,26 +44,28 @@ The following hyperparameters were used during training:
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: cosine
46
  - lr_scheduler_warmup_steps: 30
47
- - training_steps: 1400
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss |
52
  |:-------------:|:------:|:----:|:---------------:|
53
- | 0.9892 | 0.0714 | 100 | 0.9351 |
54
- | 0.7083 | 0.1429 | 200 | 0.7751 |
55
- | 0.5442 | 0.2143 | 300 | 0.7227 |
56
- | 0.4403 | 0.2857 | 400 | 0.7127 |
57
- | 0.3724 | 0.3571 | 500 | 0.7285 |
58
- | 0.3268 | 0.4286 | 600 | 0.7396 |
59
- | 0.3057 | 0.5 | 700 | 0.7587 |
60
- | 0.2798 | 0.5714 | 800 | 0.7704 |
61
- | 0.2616 | 0.6429 | 900 | 0.7888 |
62
- | 0.2418 | 0.7143 | 1000 | 0.8010 |
63
- | 0.2314 | 0.7857 | 1100 | 0.8092 |
64
- | 0.2268 | 0.8571 | 1200 | 0.8147 |
65
- | 0.223 | 0.9286 | 1300 | 0.8185 |
66
- | 0.2207 | 1.0 | 1400 | 0.8171 |
 
 
67
 
68
 
69
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [bigcode/starcoderbase-1b](https://huggingface.co/bigcode/starcoderbase-1b) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.7248
20
 
21
  ## Model description
22
 
 
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: cosine
46
  - lr_scheduler_warmup_steps: 30
47
+ - training_steps: 1600
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss |
52
  |:-------------:|:------:|:----:|:---------------:|
53
+ | 1.2597 | 0.0625 | 100 | 1.1604 |
54
+ | 0.9591 | 0.125 | 200 | 0.9402 |
55
+ | 0.8109 | 0.1875 | 300 | 0.8431 |
56
+ | 0.7151 | 0.25 | 400 | 0.7917 |
57
+ | 0.6362 | 0.3125 | 500 | 0.7607 |
58
+ | 0.5759 | 0.375 | 600 | 0.7401 |
59
+ | 0.5284 | 0.4375 | 700 | 0.7334 |
60
+ | 0.4926 | 0.5 | 800 | 0.7252 |
61
+ | 0.4616 | 0.5625 | 900 | 0.7212 |
62
+ | 0.4369 | 0.625 | 1000 | 0.7236 |
63
+ | 0.4111 | 0.6875 | 1100 | 0.7255 |
64
+ | 0.3969 | 0.75 | 1200 | 0.7236 |
65
+ | 0.3855 | 0.8125 | 1300 | 0.7260 |
66
+ | 0.3822 | 0.875 | 1400 | 0.7262 |
67
+ | 0.3768 | 0.9375 | 1500 | 0.7256 |
68
+ | 0.3778 | 1.0 | 1600 | 0.7248 |
69
 
70
 
71
  ### Framework versions