Hugging Face's logo

JackBAI
/
gpt-1l-l0-sae-1x

Model card Files Community
YX-S-Z commited on
Commit
54190d6
•
1 Parent(s): 4184603

push

Browse files
Files changed (32) hide show
  1. 0.pt +3 -0
  2. 0_cfg.json +1 -0
  3. 1.pt +3 -0
  4. 10.pt +3 -0
  5. 10_cfg.json +1 -0
  6. 11.pt +3 -0
  7. 11_cfg.json +1 -0
  8. 12.pt +3 -0
  9. 12_cfg.json +1 -0
  10. 13.pt +3 -0
  11. 13_cfg.json +1 -0
  12. 14.pt +3 -0
  13. 14_cfg.json +1 -0
  14. 15.pt +3 -0
  15. 15_cfg.json +1 -0
  16. 1_cfg.json +1 -0
  17. 2.pt +3 -0
  18. 2_cfg.json +1 -0
  19. 3.pt +3 -0
  20. 3_cfg.json +1 -0
  21. 4.pt +3 -0
  22. 4_cfg.json +1 -0
  23. 5.pt +3 -0
  24. 5_cfg.json +1 -0
  25. 6.pt +3 -0
  26. 6_cfg.json +1 -0
  27. 7.pt +3 -0
  28. 7_cfg.json +1 -0
  29. 8.pt +3 -0
  30. 8_cfg.json +1 -0
  31. 9.pt +3 -0
  32. 9_cfg.json +1 -0
0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0257a167bd43d311c051bff0a0d66c7d519c4296393bb6dfc0bead7721ca969
3
+ size 2102944
0_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fef1689cd7de0e70788145aca836a5db7fe0f0ce60af806fd6f0d658d06bdac2
3
+ size 2102944
10.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c38b8aa9b2e073b16496c7676185643060c0330d1ee05b5a8cbe91515ac3931
3
+ size 2103016
10_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
11.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a09b0205aa724f3a10d9577804fc4d68c380134eb58b185d2ec4a2192c9310c4
3
+ size 2103016
11_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
12.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:370d74b1973794b3264d10a6498d6f3b71d96d2f7b6fa2e50d92039e69a964ad
3
+ size 2103016
12_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
13.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27dbcda8339ed7ed5654a934e3c4afb2aefab7ad6863832ebbe34ecc05506303
3
+ size 2103016
13_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
14.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca1443af4ee817cdf1c29efee0a0572a0802ef070bc6237bce08c12f76410344
3
+ size 2103016
14_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
15.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6dd0729caaae795667c6104aec370f6e7e842226c91f292a976bf53100d5a37
3
+ size 2103016
15_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
1_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:090f5422580f61a463a280e1a3d06f7d9427d0be366fd0a7acbcac4e606dc6d4
3
+ size 2102944
2_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
3.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ecaf7f0b7857eefad2b9b29c7f4d941588c96fcd38b2b8dd0b6069a92f478c39
3
+ size 2102944
3_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
4.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c676adf2f212a7aca5d2c1f070762d320f30ad37e5a740adb72aec530f45d18
3
+ size 2102944
4_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
5.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3adaa86a30d405b428c086017c4bf6a103530b21f25fd2c4bb897254a29e5dd2
3
+ size 2102944
5_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
6.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d18369afea58b724687d970a438283f7e6657c6bf93bc07c42e2ee662a9c2b51
3
+ size 2102944
6_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
7.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:343a0b659485fbe4a54acb9a9b9c9c9789ed89cf908fef6a82f5701b5c3eac4a
3
+ size 2102944
7_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
8.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7c89930b1df713a63d42f745842dd15fc21b7f486d73d25ff74a975ecc064df
3
+ size 2102944
8_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}
9.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:082da55a702568098989205f4a5f0d068e0d601c5f27575b316680f839a3a3c1
3
+ size 2102944
9_cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"seed": 49, "batch_size": 4096, "buffer_mult": 384, "lr": 0.0001, "num_tokens": 2000000000, "l1_coeff": 0.0003, "beta1": 0.9, "beta2": 0.99, "dict_mult": 1, "seq_len": 128, "enc_dtype": "fp32", "remove_rare_dir": false, "model_name": "gpt-1l", "site": "post", "layer": 0, "device": "cuda:0", "model_batch_size": 512, "buffer_size": 1572864, "buffer_batches": 12288, "act_name": "blocks.0.mlp.hook_post", "act_size": 512, "dict_size": 512, "name": "gpt-1l_0_512_post"}