Hugging Face's logo

Datasets:
trl-internal-testing
/
example-images

Modalities:
Image
Formats:
imagefolder
Size:
< 1K
Libraries:
Datasets
Dataset card Data Studio Files Community
1
vwxyzjn commited on
Commit
57eb330
·
verified ·
1 Parent(s): 528ea46

Upload folder using huggingface_hub

Browse files
Files changed (8) hide show
  1. images/benchmark/pr-1540/rloo-time.pdf +0 -0
  2. images/benchmark/pr-1540/rloo-time.png +2 -2
  3. images/benchmark/pr-1540/rloo-time.svg +0 -0
  4. images/benchmark/pr-1540/rloo.pdf +0 -0
  5. images/benchmark/pr-1540/rloo.png +2 -2
  6. images/benchmark/pr-1540/rloo.svg +0 -0
  7. images/benchmark/pr-1540/rloo/models/minimal/rloo_tldr.csv +2 -2
  8. images/benchmark/pr-1540/rloo/models/minimal/rloo_tldr.md +3 -3
images/benchmark/pr-1540/rloo-time.pdf CHANGED
Binary files a/images/benchmark/pr-1540/rloo-time.pdf and b/images/benchmark/pr-1540/rloo-time.pdf differ
 
images/benchmark/pr-1540/rloo-time.png CHANGED

Git LFS Details

  • SHA256: 9770abfaedd9cd8ff5c435aa2acde1e8e0fb5bccd695a84cded4eb71d3fb6c15
  • Pointer size: 131 Bytes
  • Size of remote file: 278 kB

Git LFS Details

  • SHA256: 4d0d07971585fad30324ef229ea17350d13f6cafe98ff1a473df89a1d2b7bc97
  • Pointer size: 131 Bytes
  • Size of remote file: 268 kB
images/benchmark/pr-1540/rloo-time.svg CHANGED
images/benchmark/pr-1540/rloo.pdf CHANGED
Binary files a/images/benchmark/pr-1540/rloo.pdf and b/images/benchmark/pr-1540/rloo.pdf differ
 
images/benchmark/pr-1540/rloo.png CHANGED

Git LFS Details

  • SHA256: aa3428f4ca61cabdd3b11a8152bb9f59992be2e1d9e5e64e601658a5d6dfeb3c
  • Pointer size: 131 Bytes
  • Size of remote file: 286 kB

Git LFS Details

  • SHA256: c78b1b1d6db67031096624831dc2e5eb141dc64067cf117fb78df9a2a286a1a9
  • Pointer size: 131 Bytes
  • Size of remote file: 275 kB
images/benchmark/pr-1540/rloo.svg CHANGED
images/benchmark/pr-1540/rloo/models/minimal/rloo_tldr.csv CHANGED
@@ -1,2 +1,2 @@
1
- ,train/objective/rlhf_reward,train/objective/scores,train/objective/kl,train/objective/non_score_reward,train/objective/entropy,train/policy/approxkl_avg,train/policy/clipfrac_avg,train/loss/policy_avg,train/loss/value_avg,train/val/clipfrac_avg,train/policy/entropy_avg,train/val/ratio,train/val/ratio_var,train/val/num_eos_tokens,train/lr,train/eps
2
- huggingface/trl/cleanrl/EleutherAI_pythia-1b-deduped__sft__tldr ({'tag': ['pr-1540']}),3.8473 ± 0.0000,4.8672 ± 0.0000,20.3925 ± 0.0000,-1.0196 ± 0.0000,3.0505 ± 0.0000,0.5018 ± 0.0000,0.3758 ± 0.0000,0.1565 ± 0.0000,0.0000 ± 0.0000,0.0000 ± 0.0000,0.5307 ± 0.0000,0.9997 ± 0.0000,0.0000 ± 0.0000,2702.9500 ± 0.0000,0.0000 ± 0.0000,89.0000 ± 0.0000
 
1
+ ,train/objective/rlhf_reward,train/objective/scores,train/objective/kl,train/objective/non_score_reward,train/objective/entropy,train/policy/approxkl_avg,train/policy/clipfrac_avg,train/loss/policy_avg,train/policy/entropy_avg,train/val/ratio,train/val/ratio_var,train/val/num_eos_tokens,train/lr,train/eps
2
+ huggingface/trl/cleanrl/EleutherAI_pythia-1b-deduped__sft__tldr ({'tag': ['pr-1540']}),3.8473 ± 0.0000,4.8672 ± 0.0000,20.3925 ± 0.0000,-1.0196 ± 0.0000,3.0505 ± 0.0000,0.5018 ± 0.0000,0.3758 ± 0.0000,0.1565 ± 0.0000,0.5307 ± 0.0000,0.9997 ± 0.0000,0.0000 ± 0.0000,2702.9500 ± 0.0000,0.0000 ± 0.0000,89.0000 ± 0.0000
images/benchmark/pr-1540/rloo/models/minimal/rloo_tldr.md CHANGED
@@ -1,3 +1,3 @@
1
- | | train/objective/rlhf_reward | train/objective/scores | train/objective/kl | train/objective/non_score_reward | train/objective/entropy | train/policy/approxkl_avg | train/policy/clipfrac_avg | train/loss/policy_avg | train/loss/value_avg | train/val/clipfrac_avg | train/policy/entropy_avg | train/val/ratio | train/val/ratio_var | train/val/num_eos_tokens | train/lr | train/eps |
2
- |:---------------------------------------------------------------------------------------|:------------------------------|:-------------------------|:---------------------|:-----------------------------------|:--------------------------|:----------------------------|:----------------------------|:------------------------|:-----------------------|:-------------------------|:---------------------------|:------------------|:----------------------|:---------------------------|:----------------|:-----------------|
3
- | huggingface/trl/cleanrl/EleutherAI_pythia-1b-deduped__sft__tldr ({'tag': ['pr-1540']}) | 3.8473 ± 0.0000 | 4.8672 ± 0.0000 | 20.3925 ± 0.0000 | -1.0196 ± 0.0000 | 3.0505 ± 0.0000 | 0.5018 ± 0.0000 | 0.3758 ± 0.0000 | 0.1565 ± 0.0000 | 0.0000 ± 0.0000 | 0.0000 ± 0.0000 | 0.5307 ± 0.0000 | 0.9997 ± 0.0000 | 0.0000 ± 0.0000 | 2702.9500 ± 0.0000 | 0.0000 ± 0.0000 | 89.0000 ± 0.0000 |
 
1
+ | | train/objective/rlhf_reward | train/objective/scores | train/objective/kl | train/objective/non_score_reward | train/objective/entropy | train/policy/approxkl_avg | train/policy/clipfrac_avg | train/loss/policy_avg | train/policy/entropy_avg | train/val/ratio | train/val/ratio_var | train/val/num_eos_tokens | train/lr | train/eps |
2
+ |:---------------------------------------------------------------------------------------|:------------------------------|:-------------------------|:---------------------|:-----------------------------------|:--------------------------|:----------------------------|:----------------------------|:------------------------|:---------------------------|:------------------|:----------------------|:---------------------------|:----------------|:-----------------|
3
+ | huggingface/trl/cleanrl/EleutherAI_pythia-1b-deduped__sft__tldr ({'tag': ['pr-1540']}) | 3.8473 ± 0.0000 | 4.8672 ± 0.0000 | 20.3925 ± 0.0000 | -1.0196 ± 0.0000 | 3.0505 ± 0.0000 | 0.5018 ± 0.0000 | 0.3758 ± 0.0000 | 0.1565 ± 0.0000 | 0.5307 ± 0.0000 | 0.9997 ± 0.0000 | 0.0000 ± 0.0000 | 2702.9500 ± 0.0000 | 0.0000 ± 0.0000 | 89.0000 ± 0.0000 |