Hugging Face's logo

Datasets:
ncgc
/
influence_results

Dataset card Files
xet
 Community
ncgc commited on
Commit
a7228fd
·
verified ·
1 Parent(s): 65a5ce2

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. bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/activation_covariance.safetensors +3 -0
  2. bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/activation_eigenvalues.safetensors +3 -0
  3. bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/activation_eigenvectors.safetensors +3 -0
  4. bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/gradient_covariance.safetensors +3 -0
  5. bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/gradient_eigenvalues.safetensors +3 -0
  6. bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/gradient_eigenvectors.safetensors +3 -0
  7. bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/lambda_matrix.safetensors +3 -0
  8. bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/num_activation_covariance_processed.safetensors +3 -0
  9. bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/num_lambda_processed.safetensors +3 -0
  10. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/activation_covariance.safetensors +3 -0
  11. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/activation_eigenvalues.safetensors +3 -0
  12. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/activation_eigenvectors.safetensors +3 -0
  13. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/covariance_dataset_metadata.json +5 -0
  14. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/factor_arguments.json +20 -0
  15. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/gradient_covariance.safetensors +3 -0
  16. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/gradient_eigenvalues.safetensors +3 -0
  17. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/gradient_eigenvectors.safetensors +3 -0
  18. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/lambda_dataset_metadata.json +5 -0
  19. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/lambda_matrix.safetensors +3 -0
  20. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/num_activation_covariance_processed.safetensors +3 -0
  21. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/num_gradient_covariance_processed.safetensors +3 -0
  22. bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/num_lambda_processed.safetensors +3 -0
  23. bias_statichh-pythia-1.4b-sft-bf16/query_grads_pure_ekfac_bottom100/aggregated_query_grads.safetensors +3 -0
  24. bias_statichh-pythia-1.4b-sft-bf16/scores_ekfac_half/pairwise_scores.safetensors +3 -0
  25. bias_statichh-pythia-1.4b-sft-bf16/scores_ekfac_half/score_arguments.json +19 -0
  26. bias_statichh-pythia-1.4b-sft-bf16/scores_ekfac_half/train_dataset_metadata.json +5 -0
  27. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/activation_covariance.safetensors +3 -0
  28. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/activation_eigenvalues.safetensors +3 -0
  29. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/activation_eigenvectors.safetensors +3 -0
  30. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/covariance_dataset_metadata.json +5 -0
  31. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/factor_arguments.json +20 -0
  32. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/gradient_covariance.safetensors +3 -0
  33. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/gradient_eigenvalues.safetensors +3 -0
  34. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/gradient_eigenvectors.safetensors +3 -0
  35. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/lambda_dataset_metadata.json +5 -0
  36. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/lambda_matrix.safetensors +3 -0
  37. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/num_activation_covariance_processed.safetensors +3 -0
  38. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/num_gradient_covariance_processed.safetensors +3 -0
  39. ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/num_lambda_processed.safetensors +3 -0
  40. ethics_statichh-pythia-2.8b-sft-bf16/query_grads_pure_ekfac_bottom100/aggregated_query_grads.safetensors +3 -0
  41. ethics_statichh-pythia-2.8b-sft-bf16/scores_ekfac_half/factor_arguments.json +20 -0
  42. ethics_statichh-pythia-2.8b-sft-bf16/scores_ekfac_half/pairwise_scores.safetensors +3 -0
  43. ethics_statichh-pythia-2.8b-sft-bf16/scores_ekfac_half/query_dataset_metadata.json +5 -0
  44. ethics_statichh-pythia-2.8b-sft-bf16/scores_ekfac_half/score_arguments.json +19 -0
  45. ethics_statichh-pythia-2.8b-sft-bf16/scores_ekfac_half/train_dataset_metadata.json +5 -0
  46. ethics_statichh-pythia-6.9b-sft-bf16/factors_ekfac_half/activation_covariance.safetensors +3 -0
  47. ethics_statichh-pythia-6.9b-sft-bf16/factors_ekfac_half/factor_arguments.json +20 -0
  48. ethics_statichh-pythia-6.9b-sft-bf16/factors_ekfac_half/lambda_dataset_metadata.json +5 -0
  49. ethics_statichh-pythia-6.9b-sft-bf16/factors_ekfac_half/num_gradient_covariance_processed.safetensors +3 -0
  50. ethics_statichh-pythia-6.9b-sft-bf16/query_grads_pure_ekfac_bottom100/aggregated_query_grads.safetensors +3 -0
bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/activation_covariance.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50d380958c69141b8cebc4043c31f755727dc70d7abf3714ee54ff19e20bd0e1
3
+ size 188894582
bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/activation_eigenvalues.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:488c81765b46ef924aa7626eaee4a912111a7411f005a0116b4f6890e1010907
3
+ size 37662
bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/activation_eigenvectors.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f24c3de8432c8d2c8e1e7b4c8544b1ef512377b527a55be8a63ce9eb34872c50
3
+ size 188894582
bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/gradient_covariance.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6b46327e8a8f047b09a947b4b5c001b7c4fe8bb5047046ad550f94ebf3988e3
3
+ size 335545704
bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/gradient_eigenvalues.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9eab7389d2a4f1821e6d35bb7e144341406961c0e0a75d1f0e240c068021408f
3
+ size 47384
bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/gradient_eigenvectors.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:730df34e4810b1795865f9157c91c6061de87a3a090bf90ffcca17e6efa53074
3
+ size 335545704
bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/lambda_matrix.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4beb0d439627b229a0bab0cc64d1b6f17702df04b21c368c3c649bb041af9a7
3
+ size 93590880
bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/num_activation_covariance_processed.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04c7724eb3e45624054726f04a13a85860b92204fa3b8024b08959fadc7540ec
3
+ size 1296
bias_statichh-Qwen-1.5B-sft-bf16/factors_ekfac_half/num_lambda_processed.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14896ab6302628c478afef686cc672141ffd054f514fb44cd421316012d1390c
3
+ size 1296
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/activation_covariance.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91b009b97d3dffa84d8588a09877d83652c68fbb5f6f5ac8d57fcc9a1d54bbe7
3
+ size 159442000
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/activation_eigenvalues.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:203d0cbb7e230a6dde817b6eddea56bf8e65c4d1583fca731f52445f0c0bc524
3
+ size 29728
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/activation_eigenvectors.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab64b8ccaf9733ecc34cb6bffa9cccfdf7eae6929cf3e5dffe60b1867bc50ddf
3
+ size 159442000
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/covariance_dataset_metadata.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "type": "Dataset",
3
+ "dataset_size": 10000,
4
+ "indices": null
5
+ }
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/factor_arguments.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "strategy": "ekfac",
3
+ "use_empirical_fisher": false,
4
+ "amp_dtype": "torch.bfloat16",
5
+ "amp_scale": 65536.0,
6
+ "has_shared_parameters": false,
7
+ "covariance_max_examples": 100000,
8
+ "covariance_data_partitions": 1,
9
+ "covariance_module_partitions": 1,
10
+ "activation_covariance_dtype": "torch.bfloat16",
11
+ "gradient_covariance_dtype": "torch.bfloat16",
12
+ "eigendecomposition_dtype": "torch.float64",
13
+ "lambda_max_examples": 100000,
14
+ "lambda_data_partitions": 1,
15
+ "lambda_module_partitions": 1,
16
+ "use_iterative_lambda_aggregation": false,
17
+ "offload_activations_to_cpu": false,
18
+ "per_sample_gradient_dtype": "torch.bfloat16",
19
+ "lambda_dtype": "torch.bfloat16"
20
+ }
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/gradient_covariance.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:49bf19a9897d4760cb9c74f4467b44000da8eadcdca93f654bc0b77ec5474309
3
+ size 226493504
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/gradient_eigenvalues.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b681bb96d998e4592fd6ddd9cf135fe6cc3d7133d6269a0f396825079de74e0
3
+ size 37912
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/gradient_eigenvectors.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66cac0519dec16b2608abc1e691875265f7eb5ea2f22c9b3d9f82d91246b33f9
3
+ size 226493504
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/lambda_dataset_metadata.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "type": "Dataset",
3
+ "dataset_size": 10000,
4
+ "indices": null
5
+ }
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/lambda_matrix.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e999f16773aa7238bd794f7c4236e5c2bf4e0dffc86354424ed7057315f48f6a
3
+ size 100701248
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/num_activation_covariance_processed.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b70b9ebbcd18ed1a1a9c2a1a9e80d2129e1e0d820fba9fac771f5311e91f0b2f
3
+ size 1040
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/num_gradient_covariance_processed.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ccb9998dff91603a7af5ce203d89cc64293f794ba2275d7c48e067762621cd1
3
+ size 1040
bias_statichh-pythia-1.4b-sft-bf16/factors_ekfac_half/num_lambda_processed.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9036b194941c829309a8a501729d9bbd66e9f6b349df1fd960bdd12fb5c29ea
3
+ size 1040
bias_statichh-pythia-1.4b-sft-bf16/query_grads_pure_ekfac_bottom100/aggregated_query_grads.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c515b8050866d1757547b97abd94597abefbfc7339b9492dc37115cf351e1956
3
+ size 201400784
bias_statichh-pythia-1.4b-sft-bf16/scores_ekfac_half/pairwise_scores.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ac9d145d7ac94ea37d40926365c294d10a7340cc3298b82edef37c615b5c80c
3
+ size 193192
bias_statichh-pythia-1.4b-sft-bf16/scores_ekfac_half/score_arguments.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "damping_factor": 1e-08,
3
+ "amp_dtype": "torch.bfloat16",
4
+ "offload_activations_to_cpu": false,
5
+ "data_partitions": 1,
6
+ "module_partitions": 1,
7
+ "compute_per_module_scores": false,
8
+ "compute_per_token_scores": false,
9
+ "query_gradient_accumulation_steps": 1,
10
+ "query_gradient_low_rank": null,
11
+ "use_full_svd": false,
12
+ "aggregate_query_gradients": true,
13
+ "aggregate_train_gradients": false,
14
+ "use_measurement_for_self_influence": false,
15
+ "query_gradient_svd_dtype": "torch.float32",
16
+ "per_sample_gradient_dtype": "torch.bfloat16",
17
+ "precondition_dtype": "torch.bfloat16",
18
+ "score_dtype": "torch.bfloat16"
19
+ }
bias_statichh-pythia-1.4b-sft-bf16/scores_ekfac_half/train_dataset_metadata.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "type": "Dataset",
3
+ "dataset_size": 96256,
4
+ "indices": null
5
+ }
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/activation_covariance.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28c66d93a413acc70dbe92215f2189d670bf1535d3b52f84499a56bd2e389028
3
+ size 249109584
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/activation_eigenvalues.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dee80416261f367f641c0796e53ea53d71e527dc457b20554bb596d3cbe97e74
3
+ size 36896
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/activation_eigenvectors.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79100bef32ad44f06d9ffd9bc1b411c4ec74f67f047fe515b802d52d91ad38d8
3
+ size 249109584
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/covariance_dataset_metadata.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "type": "Dataset",
3
+ "dataset_size": 10000,
4
+ "indices": null
5
+ }
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/factor_arguments.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "strategy": "ekfac",
3
+ "use_empirical_fisher": false,
4
+ "amp_dtype": "torch.bfloat16",
5
+ "amp_scale": 65536.0,
6
+ "has_shared_parameters": false,
7
+ "covariance_max_examples": 100000,
8
+ "covariance_data_partitions": 1,
9
+ "covariance_module_partitions": 1,
10
+ "activation_covariance_dtype": "torch.bfloat16",
11
+ "gradient_covariance_dtype": "torch.bfloat16",
12
+ "eigendecomposition_dtype": "torch.float64",
13
+ "lambda_max_examples": 100000,
14
+ "lambda_data_partitions": 1,
15
+ "lambda_module_partitions": 1,
16
+ "use_iterative_lambda_aggregation": false,
17
+ "offload_activations_to_cpu": false,
18
+ "per_sample_gradient_dtype": "torch.bfloat16",
19
+ "lambda_dtype": "torch.bfloat16"
20
+ }
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/gradient_covariance.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:36cc799e5a847bd460ad4706cdd3a1dcd7914f97720001ef00b27ee7020a5546
3
+ size 353895496
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/gradient_eigenvalues.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db38ddd08af4fb6c6b644eab046db7f7d3893b5814ae5a6c39df0babb84a8ef3
3
+ size 47128
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/gradient_eigenvectors.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebd6c2429f37fbf1d44440b44b90db7378b6ba68942830f5b8b4f0f88fb21533
3
+ size 353895496
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/lambda_dataset_metadata.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "type": "Dataset",
3
+ "dataset_size": 10000,
4
+ "indices": null
5
+ }
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/lambda_matrix.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31d9d26d28772ef8bc61c482612b421919c3bc3317e3df1b25a2afa44f8c70c0
3
+ size 157333576
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/num_activation_covariance_processed.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be69209f37c9374697bb6463a826209c4c3e11d39fe4c9d87a318e870ad5c1d5
3
+ size 1040
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/num_gradient_covariance_processed.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e69185a0fdd12da566581eb45d3bcf8f9ae1dc0298e7474b10da936985bcecdd
3
+ size 1040
ethics_statichh-pythia-2.8b-sft-bf16/factors_ekfac_half/num_lambda_processed.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b43aa4a9f96cbe6c089340a64d2df866848e04e9dceb6399886e62dc6c9ea3c6
3
+ size 1040
ethics_statichh-pythia-2.8b-sft-bf16/query_grads_pure_ekfac_bottom100/aggregated_query_grads.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:162f1badd86f4b1dd7b27d2738231bc4212a9df74c3fb88cfe40f96ce908ca9a
3
+ size 314665424
ethics_statichh-pythia-2.8b-sft-bf16/scores_ekfac_half/factor_arguments.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "strategy": "ekfac",
3
+ "use_empirical_fisher": false,
4
+ "amp_dtype": "torch.bfloat16",
5
+ "amp_scale": 65536.0,
6
+ "has_shared_parameters": false,
7
+ "covariance_max_examples": 100000,
8
+ "covariance_data_partitions": 1,
9
+ "covariance_module_partitions": 1,
10
+ "activation_covariance_dtype": "torch.bfloat16",
11
+ "gradient_covariance_dtype": "torch.bfloat16",
12
+ "eigendecomposition_dtype": "torch.float64",
13
+ "lambda_max_examples": 100000,
14
+ "lambda_data_partitions": 1,
15
+ "lambda_module_partitions": 1,
16
+ "use_iterative_lambda_aggregation": false,
17
+ "offload_activations_to_cpu": false,
18
+ "per_sample_gradient_dtype": "torch.bfloat16",
19
+ "lambda_dtype": "torch.bfloat16"
20
+ }
ethics_statichh-pythia-2.8b-sft-bf16/scores_ekfac_half/pairwise_scores.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:831d5614de999bcf1f3547e51e839bb69dc24a21d842cded07369fb93e220a2f
3
+ size 193192
ethics_statichh-pythia-2.8b-sft-bf16/scores_ekfac_half/query_dataset_metadata.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "type": "Dataset",
3
+ "dataset_size": 2789,
4
+ "indices": null
5
+ }
ethics_statichh-pythia-2.8b-sft-bf16/scores_ekfac_half/score_arguments.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "damping_factor": 1e-08,
3
+ "amp_dtype": "torch.bfloat16",
4
+ "offload_activations_to_cpu": false,
5
+ "data_partitions": 1,
6
+ "module_partitions": 1,
7
+ "compute_per_module_scores": false,
8
+ "compute_per_token_scores": false,
9
+ "query_gradient_accumulation_steps": 1,
10
+ "query_gradient_low_rank": null,
11
+ "use_full_svd": false,
12
+ "aggregate_query_gradients": true,
13
+ "aggregate_train_gradients": false,
14
+ "use_measurement_for_self_influence": false,
15
+ "query_gradient_svd_dtype": "torch.float32",
16
+ "per_sample_gradient_dtype": "torch.bfloat16",
17
+ "precondition_dtype": "torch.bfloat16",
18
+ "score_dtype": "torch.bfloat16"
19
+ }
ethics_statichh-pythia-2.8b-sft-bf16/scores_ekfac_half/train_dataset_metadata.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "type": "Dataset",
3
+ "dataset_size": 96256,
4
+ "indices": null
5
+ }
ethics_statichh-pythia-6.9b-sft-bf16/factors_ekfac_half/activation_covariance.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ea48957a2ada04e4616685d5a2eb9ca77abd886fe5c50118ab6e7a8d4d59824
3
+ size 637650000
ethics_statichh-pythia-6.9b-sft-bf16/factors_ekfac_half/factor_arguments.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "strategy": "ekfac",
3
+ "use_empirical_fisher": false,
4
+ "amp_dtype": "torch.bfloat16",
5
+ "amp_scale": 65536.0,
6
+ "has_shared_parameters": false,
7
+ "covariance_max_examples": 100000,
8
+ "covariance_data_partitions": 1,
9
+ "covariance_module_partitions": 1,
10
+ "activation_covariance_dtype": "torch.bfloat16",
11
+ "gradient_covariance_dtype": "torch.bfloat16",
12
+ "eigendecomposition_dtype": "torch.float64",
13
+ "lambda_max_examples": 100000,
14
+ "lambda_data_partitions": 1,
15
+ "lambda_module_partitions": 1,
16
+ "use_iterative_lambda_aggregation": false,
17
+ "offload_activations_to_cpu": false,
18
+ "per_sample_gradient_dtype": "torch.bfloat16",
19
+ "lambda_dtype": "torch.bfloat16"
20
+ }
ethics_statichh-pythia-6.9b-sft-bf16/factors_ekfac_half/lambda_dataset_metadata.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "type": "Dataset",
3
+ "dataset_size": 10000,
4
+ "indices": null
5
+ }
ethics_statichh-pythia-6.9b-sft-bf16/factors_ekfac_half/num_gradient_covariance_processed.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f826b80c11369eedae45ee271255846aa09ab882e43a91430fcc0b3ef9a3e2a4
3
+ size 1040
ethics_statichh-pythia-6.9b-sft-bf16/query_grads_pure_ekfac_bottom100/aggregated_query_grads.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b37d98e2deb9b0ead17d171ccf52a5636033ea645882f7035b0e0edc60014f8b
3
+ size 805454288