jcandane commited on
Commit
b63b221
·
verified ·
1 Parent(s): 9b283d3

Add TGPT checkpoint

Browse files
Files changed (38) hide show
  1. checkpoint_23560/_CHECKPOINT_METADATA +1 -0
  2. checkpoint_23560/_METADATA +1 -0
  3. checkpoint_23560/_sharding +1 -0
  4. checkpoint_23560/array_metadatas/process_0 +1 -0
  5. checkpoint_23560/d/ab82cf4f9f2c82db50561e0621e480b2 +0 -0
  6. checkpoint_23560/manifest.ocdbt +0 -0
  7. checkpoint_23560/ocdbt.process_0/d/27f65457d2bfd783ea4cd938ab3e3ee4 +0 -0
  8. checkpoint_23560/ocdbt.process_0/d/29208dfdf4c901eac3dcf0422f22cf04 +0 -0
  9. checkpoint_23560/ocdbt.process_0/d/2e4be050f772f2adb35d7d4b44753c05 +0 -0
  10. checkpoint_23560/ocdbt.process_0/d/31c0e80e3c318aa7d6ac8483042c28a8 +0 -0
  11. checkpoint_23560/ocdbt.process_0/d/342cb7af1a9b3dea11573ff407e1c6fe +0 -0
  12. checkpoint_23560/ocdbt.process_0/d/393360dca0c365f0e6e075a5b64282bc +0 -0
  13. checkpoint_23560/ocdbt.process_0/d/3aecd3943f84ff0ee2d684b1a286b658 +0 -0
  14. checkpoint_23560/ocdbt.process_0/d/4312d6ca72d7aa696f98e8f29e8262ed +0 -0
  15. checkpoint_23560/ocdbt.process_0/d/4a086ea8c3ab69d710d198e6278d75eb +0 -0
  16. checkpoint_23560/ocdbt.process_0/d/4d3498cb21ea5ce7401402e2d4a47c41 +0 -0
  17. checkpoint_23560/ocdbt.process_0/d/556beac2ee97f8cccbcca7990a088870 +0 -0
  18. checkpoint_23560/ocdbt.process_0/d/5d743b1ce7d3ba391b6e0b69b983c7ab +0 -0
  19. checkpoint_23560/ocdbt.process_0/d/61bcb82ba0fe57da1d0db7c681bf543e +0 -0
  20. checkpoint_23560/ocdbt.process_0/d/6f6030b84852b12976cb5baa1857203e +0 -0
  21. checkpoint_23560/ocdbt.process_0/d/702103d08aca2628007f2cedb4287d61 +0 -0
  22. checkpoint_23560/ocdbt.process_0/d/729f961e2c8c66ee6e171d0aa26fcdf3 +0 -0
  23. checkpoint_23560/ocdbt.process_0/d/783555d7ea7cb750ef1f38d1061e04ba +0 -0
  24. checkpoint_23560/ocdbt.process_0/d/94d03fd3df8f8163cc6ac1d0b261543c +0 -0
  25. checkpoint_23560/ocdbt.process_0/d/97ab05a13eb4d155368d34be3c3aa350 +0 -0
  26. checkpoint_23560/ocdbt.process_0/d/9e13f7415daff01c16782afe977d248f +0 -0
  27. checkpoint_23560/ocdbt.process_0/d/a56095da0b4e9cc13de01f37b6770c19 +0 -0
  28. checkpoint_23560/ocdbt.process_0/d/b7c8c9921c0dfe89a59e6bda13a5b6e8 +0 -0
  29. checkpoint_23560/ocdbt.process_0/d/b841783de187dd9b3d38ed709e937222 +0 -0
  30. checkpoint_23560/ocdbt.process_0/d/cb8009e9c5803da4ac8caa37ed874b3e +0 -0
  31. checkpoint_23560/ocdbt.process_0/d/d3d5c067fd5fb4af351421077347a836 +0 -0
  32. checkpoint_23560/ocdbt.process_0/d/defddfc7fe30dd47e438e52a08775d7d +0 -0
  33. checkpoint_23560/ocdbt.process_0/d/f6d4197cff2419e8d5618f13d1065ed4 +0 -0
  34. checkpoint_23560/ocdbt.process_0/d/fa461fcb67cf094337973e50817dbabc +0 -0
  35. checkpoint_23560/ocdbt.process_0/d/fe1902870657326c59430e1bf448fb83 +0 -0
  36. checkpoint_23560/ocdbt.process_0/d/ffe4cd005c0555e517e45bcb2b05cf0b +0 -0
  37. checkpoint_23560/ocdbt.process_0/manifest.ocdbt +0 -0
  38. meta.json +1 -1
checkpoint_23560/_CHECKPOINT_METADATA ADDED
@@ -0,0 +1 @@
 
 
1
+ {"item_handlers": "orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler", "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1758137054099931399, "commit_timestamp_nsecs": 1758137054463062589, "custom_metadata": {}}
checkpoint_23560/_METADATA ADDED
@@ -0,0 +1 @@
 
 
1
+ {"tree_metadata": {"('step',)": {"key_metadata": [{"key": "step", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": []}}, "('params', 'in_proj', 'bias')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "in_proj", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'in_proj', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "in_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2, 16]}}, "('params', 'isocoder', 'AttentionBlock_0', 'Dense_0', 'bias')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64]}}, "('params', 'isocoder', 'AttentionBlock_0', 'Dense_0', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 64]}}, "('params', 'isocoder', 'AttentionBlock_0', 'Dense_1', 'bias')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'isocoder', 'AttentionBlock_0', 'Dense_1', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64, 16]}}, "('params', 'isocoder', 'AttentionBlock_0', 'LayerNorm\u03b2_0', 'bias')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "LayerNorm\u03b2_0", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'isocoder', 'AttentionBlock_0', 'LayerNorm\u03b2_0', 'scale')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "LayerNorm\u03b2_0", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'isocoder', 'AttentionBlock_0', 'LayerNorm\u03b2_1', 'bias')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "LayerNorm\u03b2_1", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'isocoder', 'AttentionBlock_0', 'LayerNorm\u03b2_1', 'scale')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "LayerNorm\u03b2_1", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'isocoder', 'AttentionBlock_0', 'MHA_0', 'k_proj', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "k_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('params', 'isocoder', 'AttentionBlock_0', 'MHA_0', 'out_proj', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "out_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('params', 'isocoder', 'AttentionBlock_0', 'MHA_0', 'q_proj', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "q_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('params', 'isocoder', 'AttentionBlock_0', 'MHA_0', 'v_proj', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "v_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('params', 'isocoder', 'AttentionBlock_1', 'Dense_0', 'bias')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64]}}, "('params', 'isocoder', 'AttentionBlock_1', 'Dense_0', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 64]}}, "('params', 'isocoder', 'AttentionBlock_1', 'Dense_1', 'bias')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'isocoder', 'AttentionBlock_1', 'Dense_1', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64, 16]}}, "('params', 'isocoder', 'AttentionBlock_1', 'LayerNorm\u03b2_0', 'bias')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "LayerNorm\u03b2_0", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'isocoder', 'AttentionBlock_1', 'LayerNorm\u03b2_0', 'scale')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "LayerNorm\u03b2_0", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'isocoder', 'AttentionBlock_1', 'LayerNorm\u03b2_1', 'bias')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "LayerNorm\u03b2_1", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'isocoder', 'AttentionBlock_1', 'LayerNorm\u03b2_1', 'scale')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "LayerNorm\u03b2_1", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'isocoder', 'AttentionBlock_1', 'MHA_0', 'k_proj', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "k_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('params', 'isocoder', 'AttentionBlock_1', 'MHA_0', 'out_proj', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "out_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('params', 'isocoder', 'AttentionBlock_1', 'MHA_0', 'q_proj', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "q_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('params', 'isocoder', 'AttentionBlock_1', 'MHA_0', 'v_proj', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "v_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('params', 'ln', 'bias')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "ln", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'ln', 'scale')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "ln", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'out_proj', 'bias')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "out_proj", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2]}}, "('params', 'out_proj', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "out_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 2]}}, "('params', 'time_proj', 'bias')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "time_proj", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('params', 'time_proj', 'kernel')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "time_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64, 16]}}, "('opt_state', '0')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "0", "key_type": 1}], "value_metadata": {"value_type": "None", "skip_deserialize": true}}, "('opt_state', '1', '0', 'count')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "count", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": []}}, "('opt_state', '1', '0', 'mu', 'in_proj', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "in_proj", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'in_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "in_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2, 16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_0', 'Dense_0', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_0', 'Dense_0', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 64]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_0', 'Dense_1', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_0', 'Dense_1', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64, 16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_0', 'LayerNorm\u03b2_0', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "LayerNorm\u03b2_0", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_0', 'LayerNorm\u03b2_0', 'scale')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "LayerNorm\u03b2_0", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_0', 'LayerNorm\u03b2_1', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "LayerNorm\u03b2_1", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_0', 'LayerNorm\u03b2_1', 'scale')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "LayerNorm\u03b2_1", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_0', 'MHA_0', 'k_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "k_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_0', 'MHA_0', 'out_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "out_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_0', 'MHA_0', 'q_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "q_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_0', 'MHA_0', 'v_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "v_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_1', 'Dense_0', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_1', 'Dense_0', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 64]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_1', 'Dense_1', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_1', 'Dense_1', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64, 16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_1', 'LayerNorm\u03b2_0', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "LayerNorm\u03b2_0", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_1', 'LayerNorm\u03b2_0', 'scale')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "LayerNorm\u03b2_0", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_1', 'LayerNorm\u03b2_1', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "LayerNorm\u03b2_1", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_1', 'LayerNorm\u03b2_1', 'scale')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "LayerNorm\u03b2_1", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_1', 'MHA_0', 'k_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "k_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_1', 'MHA_0', 'out_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "out_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_1', 'MHA_0', 'q_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "q_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'mu', 'isocoder', 'AttentionBlock_1', 'MHA_0', 'v_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "v_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'mu', 'ln', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "ln", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'ln', 'scale')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "ln", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'out_proj', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "out_proj", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2]}}, "('opt_state', '1', '0', 'mu', 'out_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "out_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 2]}}, "('opt_state', '1', '0', 'mu', 'time_proj', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "time_proj", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'mu', 'time_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "time_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64, 16]}}, "('opt_state', '1', '0', 'nu', 'in_proj', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "in_proj", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'in_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "in_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2, 16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_0', 'Dense_0', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_0', 'Dense_0', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 64]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_0', 'Dense_1', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_0', 'Dense_1', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64, 16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_0', 'LayerNorm\u03b2_0', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "LayerNorm\u03b2_0", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_0', 'LayerNorm\u03b2_0', 'scale')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "LayerNorm\u03b2_0", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_0', 'LayerNorm\u03b2_1', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "LayerNorm\u03b2_1", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_0', 'LayerNorm\u03b2_1', 'scale')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "LayerNorm\u03b2_1", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_0', 'MHA_0', 'k_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "k_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_0', 'MHA_0', 'out_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "out_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_0', 'MHA_0', 'q_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "q_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_0', 'MHA_0', 'v_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_0", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "v_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_1', 'Dense_0', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_1', 'Dense_0', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 64]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_1', 'Dense_1', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_1', 'Dense_1', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64, 16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_1', 'LayerNorm\u03b2_0', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "LayerNorm\u03b2_0", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_1', 'LayerNorm\u03b2_0', 'scale')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "LayerNorm\u03b2_0", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_1', 'LayerNorm\u03b2_1', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "LayerNorm\u03b2_1", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_1', 'LayerNorm\u03b2_1', 'scale')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "LayerNorm\u03b2_1", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_1', 'MHA_0', 'k_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "k_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_1', 'MHA_0', 'out_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "out_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_1', 'MHA_0', 'q_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "q_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'nu', 'isocoder', 'AttentionBlock_1', 'MHA_0', 'v_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "isocoder", "key_type": 2}, {"key": "AttentionBlock_1", "key_type": 2}, {"key": "MHA_0", "key_type": 2}, {"key": "v_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 16, 16]}}, "('opt_state', '1', '0', 'nu', 'ln', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "ln", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'ln', 'scale')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "ln", "key_type": 2}, {"key": "scale", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'out_proj', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "out_proj", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2]}}, "('opt_state', '1', '0', 'nu', 'out_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "out_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16, 2]}}, "('opt_state', '1', '0', 'nu', 'time_proj', 'bias')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "time_proj", "key_type": 2}, {"key": "bias", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [16]}}, "('opt_state', '1', '0', 'nu', 'time_proj', 'kernel')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "time_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [64, 16]}}, "('opt_state', '1', '1')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "1", "key_type": 1}], "value_metadata": {"value_type": "None", "skip_deserialize": true}}, "('opt_state', '1', '2')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "2", "key_type": 1}], "value_metadata": {"value_type": "None", "skip_deserialize": true}}, "('rng',)": {"key_metadata": [{"key": "rng", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [2]}}}, "use_zarr3": false, "store_array_data_equal_to_fill_value": true, "custom_metadata": null}
checkpoint_23560/_sharding ADDED
@@ -0,0 +1 @@
 
 
1
+ {"b3B0X3N0YXRlLjEuMC5jb3VudA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS50aW1lX3Byb2ouYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS50aW1lX3Byb2oua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pbl9wcm9qLmJpYXM=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pbl9wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLk1IQV8wLm91dF9wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLk1IQV8wLmtfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLk1IQV8wLnFfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLk1IQV8wLnZfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkRlbnNlXzAuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkRlbnNlXzAua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkRlbnNlXzEuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkRlbnNlXzEua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkxheWVyTm9ybc6yXzAuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkxheWVyTm9ybc6yXzAuc2NhbGU=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkxheWVyTm9ybc6yXzEuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkxheWVyTm9ybc6yXzEuc2NhbGU=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLk1IQV8wLm91dF9wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLk1IQV8wLmtfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLk1IQV8wLnFfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLk1IQV8wLnZfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkRlbnNlXzAuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkRlbnNlXzAua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkRlbnNlXzEuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkRlbnNlXzEua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkxheWVyTm9ybc6yXzAuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkxheWVyTm9ybc6yXzAuc2NhbGU=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkxheWVyTm9ybc6yXzEuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkxheWVyTm9ybc6yXzEuc2NhbGU=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5sbi5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5sbi5zY2FsZQ==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5vdXRfcHJvai5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5tdS5vdXRfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS50aW1lX3Byb2ouYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS50aW1lX3Byb2oua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pbl9wcm9qLmJpYXM=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pbl9wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLk1IQV8wLm91dF9wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLk1IQV8wLmtfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLk1IQV8wLnFfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLk1IQV8wLnZfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkRlbnNlXzAuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkRlbnNlXzAua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkRlbnNlXzEuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkRlbnNlXzEua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkxheWVyTm9ybc6yXzAuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkxheWVyTm9ybc6yXzAuc2NhbGU=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkxheWVyTm9ybc6yXzEuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18wLkxheWVyTm9ybc6yXzEuc2NhbGU=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLk1IQV8wLm91dF9wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLk1IQV8wLmtfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLk1IQV8wLnFfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLk1IQV8wLnZfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkRlbnNlXzAuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkRlbnNlXzAua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkRlbnNlXzEuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkRlbnNlXzEua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkxheWVyTm9ybc6yXzAuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkxheWVyTm9ybc6yXzAuc2NhbGU=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkxheWVyTm9ybc6yXzEuYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5pc29jb2Rlci5BdHRlbnRpb25CbG9ja18xLkxheWVyTm9ybc6yXzEuc2NhbGU=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5sbi5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5sbi5zY2FsZQ==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5vdXRfcHJvai5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","b3B0X3N0YXRlLjEuMC5udS5vdXRfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","c3RlcA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLm91dF9wcm9qLmJpYXM=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLm91dF9wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmluX3Byb2ouYmlhcw==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmluX3Byb2oua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzAuRGVuc2VfMC5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzAuRGVuc2VfMC5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzAuRGVuc2VfMS5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzAuRGVuc2VfMS5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzAuTGF5ZXJOb3JtzrJfMC5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzAuTGF5ZXJOb3JtzrJfMC5zY2FsZQ==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzAuTGF5ZXJOb3JtzrJfMS5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzAuTGF5ZXJOb3JtzrJfMS5zY2FsZQ==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzAuTUhBXzAua19wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzAuTUhBXzAub3V0X3Byb2oua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzAuTUhBXzAucV9wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzAuTUhBXzAudl9wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzEuRGVuc2VfMC5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzEuRGVuc2VfMC5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzEuRGVuc2VfMS5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzEuRGVuc2VfMS5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzEuTGF5ZXJOb3JtzrJfMC5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzEuTGF5ZXJOb3JtzrJfMC5zY2FsZQ==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzEuTGF5ZXJOb3JtzrJfMS5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzEuTGF5ZXJOb3JtzrJfMS5zY2FsZQ==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzEuTUhBXzAua19wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzEuTUhBXzAub3V0X3Byb2oua2VybmVs":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzEuTUhBXzAucV9wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmlzb2NvZGVyLkF0dGVudGlvbkJsb2NrXzEuTUhBXzAudl9wcm9qLmtlcm5lbA==":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmxuLmJpYXM=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLmxuLnNjYWxl":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLnRpbWVfcHJvai5iaWFz":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cGFyYW1zLnRpbWVfcHJvai5rZXJuZWw=":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}","cm5n":"{\"sharding_type\": \"SingleDeviceSharding\", \"device_str\": \"cuda:0\"}"}
checkpoint_23560/array_metadatas/process_0 ADDED
@@ -0,0 +1 @@
 
 
1
+ {"array_metadatas": [{"array_metadata": {"param_name": "step", "write_shape": [], "chunk_shape": [], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.in_proj.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.in_proj.kernel", "write_shape": [2, 16], "chunk_shape": [2, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_0.Dense_0.bias", "write_shape": [64], "chunk_shape": [64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_0.Dense_0.kernel", "write_shape": [16, 64], "chunk_shape": [16, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_0.Dense_1.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_0.Dense_1.kernel", "write_shape": [64, 16], "chunk_shape": [64, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_0.LayerNorm\u03b2_0.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_0.LayerNorm\u03b2_0.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_0.LayerNorm\u03b2_1.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_0.LayerNorm\u03b2_1.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_0.MHA_0.k_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_0.MHA_0.out_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_0.MHA_0.q_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_0.MHA_0.v_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_1.Dense_0.bias", "write_shape": [64], "chunk_shape": [64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_1.Dense_0.kernel", "write_shape": [16, 64], "chunk_shape": [16, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_1.Dense_1.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_1.Dense_1.kernel", "write_shape": [64, 16], "chunk_shape": [64, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_1.LayerNorm\u03b2_0.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_1.LayerNorm\u03b2_0.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_1.LayerNorm\u03b2_1.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_1.LayerNorm\u03b2_1.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_1.MHA_0.k_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_1.MHA_0.out_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_1.MHA_0.q_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.isocoder.AttentionBlock_1.MHA_0.v_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.ln.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.ln.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.out_proj.bias", "write_shape": [2], "chunk_shape": [2], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.out_proj.kernel", "write_shape": [16, 2], "chunk_shape": [16, 2], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.time_proj.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.time_proj.kernel", "write_shape": [64, 16], "chunk_shape": [64, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.count", "write_shape": [], "chunk_shape": [], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.in_proj.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.in_proj.kernel", "write_shape": [2, 16], "chunk_shape": [2, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_0.Dense_0.bias", "write_shape": [64], "chunk_shape": [64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_0.Dense_0.kernel", "write_shape": [16, 64], "chunk_shape": [16, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_0.Dense_1.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_0.Dense_1.kernel", "write_shape": [64, 16], "chunk_shape": [64, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_0.LayerNorm\u03b2_0.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_0.LayerNorm\u03b2_0.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_0.LayerNorm\u03b2_1.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_0.LayerNorm\u03b2_1.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_0.MHA_0.k_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_0.MHA_0.out_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_0.MHA_0.q_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_0.MHA_0.v_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_1.Dense_0.bias", "write_shape": [64], "chunk_shape": [64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_1.Dense_0.kernel", "write_shape": [16, 64], "chunk_shape": [16, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_1.Dense_1.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_1.Dense_1.kernel", "write_shape": [64, 16], "chunk_shape": [64, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_1.LayerNorm\u03b2_0.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_1.LayerNorm\u03b2_0.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_1.LayerNorm\u03b2_1.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_1.LayerNorm\u03b2_1.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_1.MHA_0.k_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_1.MHA_0.out_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_1.MHA_0.q_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.isocoder.AttentionBlock_1.MHA_0.v_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.ln.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.ln.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.out_proj.bias", "write_shape": [2], "chunk_shape": [2], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.out_proj.kernel", "write_shape": [16, 2], "chunk_shape": [16, 2], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.time_proj.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.time_proj.kernel", "write_shape": [64, 16], "chunk_shape": [64, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.in_proj.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.in_proj.kernel", "write_shape": [2, 16], "chunk_shape": [2, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_0.Dense_0.bias", "write_shape": [64], "chunk_shape": [64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_0.Dense_0.kernel", "write_shape": [16, 64], "chunk_shape": [16, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_0.Dense_1.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_0.Dense_1.kernel", "write_shape": [64, 16], "chunk_shape": [64, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_0.LayerNorm\u03b2_0.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_0.LayerNorm\u03b2_0.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_0.LayerNorm\u03b2_1.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_0.LayerNorm\u03b2_1.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_0.MHA_0.k_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_0.MHA_0.out_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_0.MHA_0.q_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_0.MHA_0.v_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_1.Dense_0.bias", "write_shape": [64], "chunk_shape": [64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_1.Dense_0.kernel", "write_shape": [16, 64], "chunk_shape": [16, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_1.Dense_1.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_1.Dense_1.kernel", "write_shape": [64, 16], "chunk_shape": [64, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_1.LayerNorm\u03b2_0.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_1.LayerNorm\u03b2_0.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_1.LayerNorm\u03b2_1.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_1.LayerNorm\u03b2_1.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_1.MHA_0.k_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_1.MHA_0.out_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_1.MHA_0.q_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.isocoder.AttentionBlock_1.MHA_0.v_proj.kernel", "write_shape": [1, 16, 16], "chunk_shape": [1, 16, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.ln.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.ln.scale", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.out_proj.bias", "write_shape": [2], "chunk_shape": [2], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.out_proj.kernel", "write_shape": [16, 2], "chunk_shape": [16, 2], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.time_proj.bias", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.time_proj.kernel", "write_shape": [64, 16], "chunk_shape": [64, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "rng", "write_shape": [2], "chunk_shape": [2], "ext_metadata": null}}]}
checkpoint_23560/d/ab82cf4f9f2c82db50561e0621e480b2 ADDED
Binary file (26.1 kB). View file
 
checkpoint_23560/manifest.ocdbt ADDED
Binary file (118 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/27f65457d2bfd783ea4cd938ab3e3ee4 ADDED
Binary file (72.5 kB). View file
 
checkpoint_23560/ocdbt.process_0/d/29208dfdf4c901eac3dcf0422f22cf04 ADDED
Binary file (628 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/2e4be050f772f2adb35d7d4b44753c05 ADDED
Binary file (646 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/31c0e80e3c318aa7d6ac8483042c28a8 ADDED
Binary file (657 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/342cb7af1a9b3dea11573ff407e1c6fe ADDED
Binary file (626 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/393360dca0c365f0e6e075a5b64282bc ADDED
Binary file (611 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/3aecd3943f84ff0ee2d684b1a286b658 ADDED
Binary file (647 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/4312d6ca72d7aa696f98e8f29e8262ed ADDED
Binary file (611 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/4a086ea8c3ab69d710d198e6278d75eb ADDED
Binary file (650 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/4d3498cb21ea5ce7401402e2d4a47c41 ADDED
Binary file (650 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/556beac2ee97f8cccbcca7990a088870 ADDED
Binary file (451 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/5d743b1ce7d3ba391b6e0b69b983c7ab ADDED
Binary file (656 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/61bcb82ba0fe57da1d0db7c681bf543e ADDED
Binary file (610 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/6f6030b84852b12976cb5baa1857203e ADDED
Binary file (619 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/702103d08aca2628007f2cedb4287d61 ADDED
Binary file (621 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/729f961e2c8c66ee6e171d0aa26fcdf3 ADDED
Binary file (686 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/783555d7ea7cb750ef1f38d1061e04ba ADDED
Binary file (4.51 kB). View file
 
checkpoint_23560/ocdbt.process_0/d/94d03fd3df8f8163cc6ac1d0b261543c ADDED
Binary file (579 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/97ab05a13eb4d155368d34be3c3aa350 ADDED
Binary file (17.9 kB). View file
 
checkpoint_23560/ocdbt.process_0/d/9e13f7415daff01c16782afe977d248f ADDED
Binary file (640 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/a56095da0b4e9cc13de01f37b6770c19 ADDED
Binary file (685 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/b7c8c9921c0dfe89a59e6bda13a5b6e8 ADDED
Binary file (619 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/b841783de187dd9b3d38ed709e937222 ADDED
Binary file (654 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/cb8009e9c5803da4ac8caa37ed874b3e ADDED
Binary file (626 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/d3d5c067fd5fb4af351421077347a836 ADDED
Binary file (510 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/defddfc7fe30dd47e438e52a08775d7d ADDED
Binary file (676 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/f6d4197cff2419e8d5618f13d1065ed4 ADDED
Binary file (171 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/fa461fcb67cf094337973e50817dbabc ADDED
Binary file (674 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/fe1902870657326c59430e1bf448fb83 ADDED
Binary file (450 Bytes). View file
 
checkpoint_23560/ocdbt.process_0/d/ffe4cd005c0555e517e45bcb2b05cf0b ADDED
Binary file (1.32 kB). View file
 
checkpoint_23560/ocdbt.process_0/manifest.ocdbt ADDED
Binary file (608 Bytes). View file
 
meta.json CHANGED
@@ -6,7 +6,7 @@
6
  "d_model": 16,
7
  "n_heads": 1,
8
  "depth": 2,
9
- "\u03b2": 0.1,
10
  "predict_delta": true,
11
  "seasonal_lag": null,
12
  "out_mode": "mse",
 
6
  "d_model": 16,
7
  "n_heads": 1,
8
  "depth": 2,
9
+ "\u03b2": 0.01,
10
  "predict_delta": true,
11
  "seasonal_lag": null,
12
  "out_mode": "mse",