Hugging Face's logo

AlexWortega
/
openjev

Text Classification
Transformers
Safetensors
English
nli
cross-encoder
qwen3.5
reranker
image-text-to-text
Model card Files
xet
 Community
5
AlexWortega commited on
Commit
a07c166
·
verified ·
1 Parent(s): 8c9db06

35B latent+MLP heads

Browse files
Files changed (27) hide show
  1. mlp_heads_35b/arc_challenge/head.pt +3 -0
  2. mlp_heads_35b/arc_challenge/meta.json +9 -0
  3. mlp_heads_35b/arc_challenge/norm.npz +3 -0
  4. mlp_heads_35b/arc_easy/head.pt +3 -0
  5. mlp_heads_35b/arc_easy/meta.json +9 -0
  6. mlp_heads_35b/arc_easy/norm.npz +3 -0
  7. mlp_heads_35b/chess/head.pt +3 -0
  8. mlp_heads_35b/chess/meta.json +9 -0
  9. mlp_heads_35b/chess/norm.npz +3 -0
  10. mlp_heads_35b/gpqa/head.pt +3 -0
  11. mlp_heads_35b/gpqa/meta.json +9 -0
  12. mlp_heads_35b/gpqa/norm.npz +3 -0
  13. mlp_heads_35b/gsm8k_mc10/head.pt +3 -0
  14. mlp_heads_35b/gsm8k_mc10/meta.json +9 -0
  15. mlp_heads_35b/gsm8k_mc10/norm.npz +3 -0
  16. mlp_heads_35b/gsm8k_mc4/head.pt +3 -0
  17. mlp_heads_35b/gsm8k_mc4/meta.json +9 -0
  18. mlp_heads_35b/gsm8k_mc4/norm.npz +3 -0
  19. mlp_heads_35b/hellaswag/head.pt +3 -0
  20. mlp_heads_35b/hellaswag/meta.json +9 -0
  21. mlp_heads_35b/hellaswag/norm.npz +3 -0
  22. mlp_heads_35b/mmlu/head.pt +3 -0
  23. mlp_heads_35b/mmlu/meta.json +9 -0
  24. mlp_heads_35b/mmlu/norm.npz +3 -0
  25. mlp_heads_35b/winogrande/head.pt +3 -0
  26. mlp_heads_35b/winogrande/meta.json +9 -0
  27. mlp_heads_35b/winogrande/norm.npz +3 -0
mlp_heads_35b/arc_challenge/head.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:908ebe7c9a02476b34d5bff8dcc8cf51c10e6de1726775ddd165d3045d2c4f3a
3
+ size 4200767
mlp_heads_35b/arc_challenge/meta.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "task": "arc_challenge",
3
+ "d": 2048,
4
+ "hidden": 512,
5
+ "dropout": 0.1,
6
+ "eps": 0.1,
7
+ "test_acc": 0.9172354948805461,
8
+ "hypothesis": "The correct answer is: {option}"
9
+ }
mlp_heads_35b/arc_challenge/norm.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e93215f897747bdd6a1efd26e4a856468ddc2886dbbc98c9f4e05e93eb1a124
3
+ size 16878
mlp_heads_35b/arc_easy/head.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f16b9feb3bf489fae52a208bed4bbf84c3d869a6aced22587a9cb08c2052c4f
3
+ size 4200767
mlp_heads_35b/arc_easy/meta.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "task": "arc_easy",
3
+ "d": 2048,
4
+ "hidden": 512,
5
+ "dropout": 0.1,
6
+ "eps": 0.1,
7
+ "test_acc": 0.9654882154882155,
8
+ "hypothesis": "The correct answer is: {option}"
9
+ }
mlp_heads_35b/arc_easy/norm.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26f179c94a9e1830991a63b3905801b0b130a742bbdafcdd9302a8688d1eaa2a
3
+ size 16878
mlp_heads_35b/chess/head.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c319784a2fe198135ae3aa1c927f6accd09b6e118dc14de14024ad6b4e7461f
3
+ size 4200767
mlp_heads_35b/chess/meta.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "task": "chess",
3
+ "d": 2048,
4
+ "hidden": 512,
5
+ "dropout": 0.1,
6
+ "eps": 0.1,
7
+ "test_acc": 0.896,
8
+ "hypothesis": "The correct answer is: {option}"
9
+ }
mlp_heads_35b/chess/norm.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9e9ea2333c75230e335f5e72e652d8f17d31d8e3cafeebc6e10293bc30f0846
3
+ size 16878
mlp_heads_35b/gpqa/head.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6361aaee1db7e4770994dcc6e7aa4644d2dae8d4e036e0585f8c11b153f3f136
3
+ size 4200767
mlp_heads_35b/gpqa/meta.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "task": "gpqa",
3
+ "d": 2048,
4
+ "hidden": 512,
5
+ "dropout": 0.1,
6
+ "eps": 0.1,
7
+ "test_acc": 0.3838383838383838,
8
+ "hypothesis": "The correct answer is: {option}"
9
+ }
mlp_heads_35b/gpqa/norm.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e124adf6b0fa5e33a258c0a2a84831f2b06694c1bcf90d4ae245d54c488f0d0
3
+ size 16878
mlp_heads_35b/gsm8k_mc10/head.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14f789a07497ce1b61218abf05544f9a46ad1e54e32f0b48794657ee708cb56e
3
+ size 4200767
mlp_heads_35b/gsm8k_mc10/meta.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "task": "gsm8k_mc10",
3
+ "d": 2048,
4
+ "hidden": 512,
5
+ "dropout": 0.1,
6
+ "eps": 0.1,
7
+ "test_acc": 0.6034874905231236,
8
+ "hypothesis": "The correct answer is: {option}"
9
+ }
mlp_heads_35b/gsm8k_mc10/norm.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:711526270655e1dc23aab86f1072a41f23d7baeb1d1e30e24bae6bc1bf8381fe
3
+ size 16878
mlp_heads_35b/gsm8k_mc4/head.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9854dd09f4c0e197bbd0b242609330538e63f62a0e05be65ef5eab61e16989f9
3
+ size 4200767
mlp_heads_35b/gsm8k_mc4/meta.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "task": "gsm8k_mc4",
3
+ "d": 2048,
4
+ "hidden": 512,
5
+ "dropout": 0.1,
6
+ "eps": 0.1,
7
+ "test_acc": 0.7232752084912812,
8
+ "hypothesis": "The correct answer is: {option}"
9
+ }
mlp_heads_35b/gsm8k_mc4/norm.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f247a46985971895d4e242832285766f23fe1c8ca9f7b4bb87bb1d1d1bf490ab
3
+ size 16878
mlp_heads_35b/hellaswag/head.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b45ee6219018ab4ffbec5107f4f8708be83b78f453102c7b20a99b6b2dbdac46
3
+ size 4200767
mlp_heads_35b/hellaswag/meta.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "task": "hellaswag",
3
+ "d": 2048,
4
+ "hidden": 512,
5
+ "dropout": 0.1,
6
+ "eps": 0.1,
7
+ "test_acc": 0.9428400716988647,
8
+ "hypothesis": "The correct answer is: {option}"
9
+ }
mlp_heads_35b/hellaswag/norm.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5a0038ee638e197fdd67e2bd7c13083faa3cc2516cfe88609b7bc919b88043c
3
+ size 16878
mlp_heads_35b/mmlu/head.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:921abd26f9b2a97fe8c4fe022e1b98b62187f0a6bb6a56c5e164dfb54220948a
3
+ size 4200767
mlp_heads_35b/mmlu/meta.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "task": "mmlu",
3
+ "d": 2048,
4
+ "hidden": 512,
5
+ "dropout": 0.1,
6
+ "eps": 0.1,
7
+ "test_acc": 0.7238997293832787,
8
+ "hypothesis": "The correct answer is: {option}"
9
+ }
mlp_heads_35b/mmlu/norm.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b6e4ed2904a5a5c2935d5bd33a56cde24b472329c18f664db0ef89e7c45e88f
3
+ size 16878
mlp_heads_35b/winogrande/head.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fcfddaf0d53a380f868449bfc899c0d6040015315c8d8896a64ad7b54d334933
3
+ size 4200767
mlp_heads_35b/winogrande/meta.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "task": "winogrande",
3
+ "d": 2048,
4
+ "hidden": 512,
5
+ "dropout": 0.1,
6
+ "eps": 0.1,
7
+ "test_acc": 0.8121546961325967,
8
+ "hypothesis": "The correct answer is: {option}"
9
+ }
mlp_heads_35b/winogrande/norm.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f1d502ef702fb3a45ba30463fbb1edec0556d73508631711fdda98235879a71
3
+ size 16878