Add TGPT checkpoint
Browse files- checkpoint_24800/_CHECKPOINT_METADATA +1 -0
- checkpoint_24800/_METADATA +1 -0
- checkpoint_24800/_sharding +1 -0
- checkpoint_24800/array_metadatas/process_0 +1 -0
- checkpoint_24800/d/90c6d68e8c73e23227625dd2f577c2be +0 -0
- checkpoint_24800/manifest.ocdbt +0 -0
- checkpoint_24800/ocdbt.process_0/d/12c0b7cbe597f7124a45416d5867a7bc +0 -0
- checkpoint_24800/ocdbt.process_0/d/15b9cda0d6f89153bbb769de26824d92 +0 -0
- checkpoint_24800/ocdbt.process_0/d/176a432181bab1aceb04db366818e9e4 +0 -0
- checkpoint_24800/ocdbt.process_0/d/179cf1d082d96d2f6a3975aaf632c3d7 +0 -0
- checkpoint_24800/ocdbt.process_0/d/2bb87d3d5da28e97def0b035d81b6c73 +0 -0
- checkpoint_24800/ocdbt.process_0/d/3e1b05fdb65df07132da733e4e16fd08 +0 -0
- checkpoint_24800/ocdbt.process_0/d/44864593cd4797f8d414555486fcecb8 +0 -0
- checkpoint_24800/ocdbt.process_0/d/4aaccbd95a6bd8a64755ed2aa639ce16 +0 -0
- checkpoint_24800/ocdbt.process_0/d/4c42e63b82a982fe248daf4fb9d776ca +0 -0
- checkpoint_24800/ocdbt.process_0/d/4e042ebce2847ba553444851c9d84cfa +0 -0
- checkpoint_24800/ocdbt.process_0/d/4e80689478505d8abee6a25e906c28b3 +0 -0
- checkpoint_24800/ocdbt.process_0/d/4fd9075a9dbe5c21aef1702548f44a0d +0 -0
- checkpoint_24800/ocdbt.process_0/d/5ed2d507777416ed7ae96f44742c8763 +0 -0
- checkpoint_24800/ocdbt.process_0/d/66afd6f34bb5b67436163fc1f9d79b2d +0 -0
- checkpoint_24800/ocdbt.process_0/d/69a44e4556d16d0ba6e88c7960855fe1 +0 -0
- checkpoint_24800/ocdbt.process_0/d/6a775a631a16e488770b2e8bb7a33b4b +0 -0
- checkpoint_24800/ocdbt.process_0/d/7129d111b7ff730f96960ea145a2f512 +0 -0
- checkpoint_24800/ocdbt.process_0/d/7f6101bf36ac3646a275981e560f5198 +0 -0
- checkpoint_24800/ocdbt.process_0/d/7fa73ee68dc03125f32d4566f9764e41 +0 -0
- checkpoint_24800/ocdbt.process_0/d/8e013befed710544f833b75d5cfbae28 +0 -0
- checkpoint_24800/ocdbt.process_0/d/97f65ff3bc361c6390d8515e1c600f49 +0 -0
- checkpoint_24800/ocdbt.process_0/d/a16fe455db0a9e50569071e231da7203 +0 -0
- checkpoint_24800/ocdbt.process_0/d/a9f464a218ca773ffbeaad19c8a16444 +0 -0
- checkpoint_24800/ocdbt.process_0/d/aad5c2910de2384bdc91cec94187b8fd +0 -0
- checkpoint_24800/ocdbt.process_0/d/b5cf0aea832fc615d9cb733e66a77c81 +0 -0
- checkpoint_24800/ocdbt.process_0/d/be7af4339497af1b6ae6e3339cc053c9 +0 -0
- checkpoint_24800/ocdbt.process_0/d/cfd0fda7c6f3c3031e48b21df064e91c +0 -0
- checkpoint_24800/ocdbt.process_0/d/d10eea598e8a99b5bfc3a01fec5d4e9f +0 -0
- checkpoint_24800/ocdbt.process_0/d/dece58c35268206ceabd007122668711 +0 -0
- checkpoint_24800/ocdbt.process_0/d/e5f4dea2f4814c0ac6a4e3b9e946f235 +0 -0
- checkpoint_24800/ocdbt.process_0/d/f1fb4b8599e50cf1202e15b2f2e6ee12 +0 -0
- checkpoint_24800/ocdbt.process_0/d/fecdc64ab85d151bf2fa95c3c08f3da3 +0 -0
- checkpoint_24800/ocdbt.process_0/manifest.ocdbt +0 -0
- meta.json +1 -1
checkpoint_24800/_CHECKPOINT_METADATA
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"item_handlers": "orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler", "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1758137161157552878, "commit_timestamp_nsecs": 1758137161464742372, "custom_metadata": {}}
|
checkpoint_24800/_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_24800/_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_24800/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_24800/d/90c6d68e8c73e23227625dd2f577c2be
ADDED
|
Binary file (27.1 kB). View file
|
|
|
checkpoint_24800/manifest.ocdbt
ADDED
|
Binary file (118 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/12c0b7cbe597f7124a45416d5867a7bc
ADDED
|
Binary file (11.4 kB). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/15b9cda0d6f89153bbb769de26824d92
ADDED
|
Binary file (646 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/176a432181bab1aceb04db366818e9e4
ADDED
|
Binary file (31.2 kB). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/179cf1d082d96d2f6a3975aaf632c3d7
ADDED
|
Binary file (697 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/2bb87d3d5da28e97def0b035d81b6c73
ADDED
|
Binary file (679 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/3e1b05fdb65df07132da733e4e16fd08
ADDED
|
Binary file (625 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/44864593cd4797f8d414555486fcecb8
ADDED
|
Binary file (610 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/4aaccbd95a6bd8a64755ed2aa639ce16
ADDED
|
Binary file (623 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/4c42e63b82a982fe248daf4fb9d776ca
ADDED
|
Binary file (645 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/4e042ebce2847ba553444851c9d84cfa
ADDED
|
Binary file (686 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/4e80689478505d8abee6a25e906c28b3
ADDED
|
Binary file (650 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/4fd9075a9dbe5c21aef1702548f44a0d
ADDED
|
Binary file (685 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/5ed2d507777416ed7ae96f44742c8763
ADDED
|
Binary file (611 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/66afd6f34bb5b67436163fc1f9d79b2d
ADDED
|
Binary file (632 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/69a44e4556d16d0ba6e88c7960855fe1
ADDED
|
Binary file (650 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/6a775a631a16e488770b2e8bb7a33b4b
ADDED
|
Binary file (618 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/7129d111b7ff730f96960ea145a2f512
ADDED
|
Binary file (625 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/7f6101bf36ac3646a275981e560f5198
ADDED
|
Binary file (662 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/7fa73ee68dc03125f32d4566f9764e41
ADDED
|
Binary file (626 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/8e013befed710544f833b75d5cfbae28
ADDED
|
Binary file (533 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/97f65ff3bc361c6390d8515e1c600f49
ADDED
|
Binary file (661 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/a16fe455db0a9e50569071e231da7203
ADDED
|
Binary file (650 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/a9f464a218ca773ffbeaad19c8a16444
ADDED
|
Binary file (171 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/aad5c2910de2384bdc91cec94187b8fd
ADDED
|
Binary file (1.34 kB). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/b5cf0aea832fc615d9cb733e66a77c81
ADDED
|
Binary file (657 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/be7af4339497af1b6ae6e3339cc053c9
ADDED
|
Binary file (657 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/cfd0fda7c6f3c3031e48b21df064e91c
ADDED
|
Binary file (681 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/d10eea598e8a99b5bfc3a01fec5d4e9f
ADDED
|
Binary file (59.8 kB). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/dece58c35268206ceabd007122668711
ADDED
|
Binary file (640 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/e5f4dea2f4814c0ac6a4e3b9e946f235
ADDED
|
Binary file (452 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/f1fb4b8599e50cf1202e15b2f2e6ee12
ADDED
|
Binary file (610 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/d/fecdc64ab85d151bf2fa95c3c08f3da3
ADDED
|
Binary file (629 Bytes). View file
|
|
|
checkpoint_24800/ocdbt.process_0/manifest.ocdbt
ADDED
|
Binary file (669 Bytes). View file
|
|
|
meta.json
CHANGED
|
@@ -6,7 +6,7 @@
|
|
| 6 |
"d_model": 16,
|
| 7 |
"n_heads": 1,
|
| 8 |
"depth": 2,
|
| 9 |
-
"\u03b2": 0.
|
| 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": 0.005,
|
| 10 |
"predict_delta": true,
|
| 11 |
"seasonal_lag": null,
|
| 12 |
"out_mode": "mse",
|