Hugging Face's logo

Datasets:
Exploration-Lab
/
dim-discovery-archive

ArXiv:
Dataset card Files
xet
 Community
Divyanshu-Bhatt commited on
Commit
30d8e0b
·
verified ·
1 Parent(s): 9ddccae

Add files using upload-large-folder tool

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.0.hook_mlp_out.json +1 -0
  2. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.11.hook_mlp_out.json +1 -0
  3. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.12.hook_mlp_out.json +1 -0
  4. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.13.hook_mlp_out.json +1 -0
  5. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.13.hook_resid_post.json +1 -0
  6. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.14.hook_mlp_out.json +1 -0
  7. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.16.hook_mlp_out.json +1 -0
  8. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.16.hook_resid_post.json +1 -0
  9. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.17.hook_mlp_out.json +1 -0
  10. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.19.hook_resid_post.json +1 -0
  11. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.2.hook_mlp_out.json +1 -0
  12. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.2.hook_resid_post.json +1 -0
  13. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.20.hook_mlp_out.json +1 -0
  14. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.21.hook_resid_post.json +1 -0
  15. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.22.hook_mlp_out.json +1 -0
  16. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.23.hook_mlp_out.json +1 -0
  17. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.24.hook_resid_post.json +1 -0
  18. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.25.hook_mlp_out.json +1 -0
  19. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.26.hook_mlp_out.json +1 -0
  20. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.27.hook_mlp_out.json +1 -0
  21. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.3.hook_mlp_out.json +1 -0
  22. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.3.hook_resid_post.json +1 -0
  23. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.32.hook_resid_post.json +1 -0
  24. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.33.hook_resid_post.json +1 -0
  25. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.36.hook_resid_post.json +1 -0
  26. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.37.hook_resid_post.json +1 -0
  27. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.38.hook_resid_post.json +1 -0
  28. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.39.hook_mlp_out.json +1 -0
  29. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.39.hook_resid_post.json +1 -0
  30. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.40.hook_mlp_out.json +1 -0
  31. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.40.hook_resid_post.json +1 -0
  32. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.41.hook_mlp_out.json +1 -0
  33. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.41.hook_resid_post.json +1 -0
  34. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.43.hook_mlp_out.json +1 -0
  35. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.44.hook_mlp_out.json +1 -0
  36. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.44.hook_resid_post.json +1 -0
  37. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.45.hook_mlp_out.json +1 -0
  38. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.45.hook_resid_post.json +1 -0
  39. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.46.hook_mlp_out.json +1 -0
  40. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.47.hook_mlp_out.json +1 -0
  41. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.5.hook_mlp_out.json +1 -0
  42. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.6.hook_mlp_out.json +1 -0
  43. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.6.hook_resid_post.json +1 -0
  44. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.7.hook_mlp_out.json +1 -0
  45. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.7.hook_resid_post.json +1 -0
  46. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.8.hook_resid_post.json +1 -0
  47. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.9.hook_resid_post.json +1 -0
  48. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_ln_final.hook_normalized.json +1 -0
  49. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_output.json +1 -0
  50. sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/intrinsic/mle_local_dims/blocks.0.hook_resid_post.csv +0 -0
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.0.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.11.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.12.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 1, "1": 444}, "net_accuracy": 0.5103211009174312}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.13.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 188, "1": 290}, "net_accuracy": 0.5481651376146789}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.13.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.14.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 0, "1": 444}, "net_accuracy": 0.5091743119266054}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.16.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.16.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.17.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 0, "1": 444}, "net_accuracy": 0.5091743119266054}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.19.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.2.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.2.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.20.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 3, "1": 441}, "net_accuracy": 0.5091743119266054}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.21.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.22.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 2}, "net_accuracy": 0.49311926605504586}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.23.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 415, "1": 10}, "net_accuracy": 0.4873853211009174}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.24.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.25.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 401, "1": 10}, "net_accuracy": 0.4713302752293578}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.26.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 81, "1": 362}, "net_accuracy": 0.5080275229357798}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.27.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 0, "1": 444}, "net_accuracy": 0.5091743119266054}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.3.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 299, "1": 145}, "net_accuracy": 0.5091743119266054}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.3.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.32.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.33.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.36.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.37.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.38.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.39.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 221, "1": 164}, "net_accuracy": 0.4415137614678899}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.39.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.40.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 3, "1": 444}, "net_accuracy": 0.5126146788990825}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.40.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.41.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 216, "1": 207}, "net_accuracy": 0.48509174311926606}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.41.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.43.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 0, "1": 444}, "net_accuracy": 0.5091743119266054}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.44.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.44.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.45.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.45.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.46.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 0, "1": 444}, "net_accuracy": 0.5091743119266054}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.47.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 9, "1": 420}, "net_accuracy": 0.4919724770642202}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.5.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.6.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.6.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.7.hook_mlp_out.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 45, "1": 403}, "net_accuracy": 0.5137614678899083}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.7.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.8.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_blocks.9.hook_resid_post.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 0}, "net_accuracy": 0.4908256880733945}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_ln_final.hook_normalized.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 1}, "net_accuracy": 0.4919724770642202}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/accuracy_output.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total_per_class": {"0": 428, "1": 444}, "correct_per_class": {"0": 428, "1": 1}, "net_accuracy": 0.4919724770642202}
sst2_validation/few_shot_0/gpt2-xl/last_token_relative_depth_100/intrinsic/mle_local_dims/blocks.0.hook_resid_post.csv ADDED
The diff for this file is too large to render. See raw diff