Add TGPT checkpoint
Browse files- checkpoint_18600/_CHECKPOINT_METADATA +1 -0
- checkpoint_18600/_METADATA +1 -0
- checkpoint_18600/_sharding +1 -0
- checkpoint_18600/array_metadatas/process_0 +1 -0
- checkpoint_18600/d/e8fb8e2162221b8b4302e5b519d00d31 +0 -0
- checkpoint_18600/manifest.ocdbt +0 -0
- checkpoint_18600/ocdbt.process_0/d/0269e8247fae2ea597c436f321b5869e +0 -0
- checkpoint_18600/ocdbt.process_0/d/0c7c1ae583522307260686d6c3a7c0b7 +0 -0
- checkpoint_18600/ocdbt.process_0/d/1a3f971051c91416d9e759fff379296b +0 -0
- checkpoint_18600/ocdbt.process_0/d/378945e046b553d80898ea302983eeca +0 -0
- checkpoint_18600/ocdbt.process_0/d/39a4c57f34c8a68743f73a6791a4d74a +0 -0
- checkpoint_18600/ocdbt.process_0/d/43139bbffdfc30d4c68eb9d62692f024 +0 -0
- checkpoint_18600/ocdbt.process_0/d/46097bc2711f3f540658dcda7dde68b7 +0 -0
- checkpoint_18600/ocdbt.process_0/d/48589a405f7063fe1c8658c17fbfa08b +0 -0
- checkpoint_18600/ocdbt.process_0/d/4a43a24736782dc9c8343725bb54c427 +0 -0
- checkpoint_18600/ocdbt.process_0/d/6e0dbd3904372f423da34b39e52374c5 +0 -0
- checkpoint_18600/ocdbt.process_0/d/759214f7539ef1c601912593389dabba +0 -0
- checkpoint_18600/ocdbt.process_0/d/851d172ed5a8bebf5738800f9b928182 +0 -0
- checkpoint_18600/ocdbt.process_0/d/88a52735f04097dc8c7297712cbebdea +0 -0
- checkpoint_18600/ocdbt.process_0/d/8f6d622c66c039fbbe7388bd91cd82b9 +0 -0
- checkpoint_18600/ocdbt.process_0/d/93882eade5f9788c7a685306c6189e4c +0 -0
- checkpoint_18600/ocdbt.process_0/d/9a33d4987f10544ce76c12a17b1cc16d +0 -0
- checkpoint_18600/ocdbt.process_0/d/9d3d87aa350567a46666b19e15adf058 +0 -0
- checkpoint_18600/ocdbt.process_0/d/a599a292c89dc5c53f9af91172b5bd12 +0 -0
- checkpoint_18600/ocdbt.process_0/d/a7ae19bf629bfeb4040f4d8b59573f89 +0 -0
- checkpoint_18600/ocdbt.process_0/d/a89624385b8088f672240bfea3b86939 +0 -0
- checkpoint_18600/ocdbt.process_0/d/b0992e85bfe7f52b028c582a4619aa9f +0 -0
- checkpoint_18600/ocdbt.process_0/d/ba739de292ec6408bbe0dfe5d1148fb5 +0 -0
- checkpoint_18600/ocdbt.process_0/d/c63ceb0c1c72dd9574975e00b15917c2 +0 -0
- checkpoint_18600/ocdbt.process_0/d/c85fd8ff0a933cc6f574b875d0d985c3 +0 -0
- checkpoint_18600/ocdbt.process_0/d/d30e7cfe83ed986597db688d5d6b457d +0 -0
- checkpoint_18600/ocdbt.process_0/d/d39f9542b8b18c235f96e367a8f07085 +0 -0
- checkpoint_18600/ocdbt.process_0/d/ea9aef7a542d893d475aae18de959dc0 +0 -0
- checkpoint_18600/ocdbt.process_0/d/eb257f9ab066115b62b3df3f7b2c1c74 +0 -0
- checkpoint_18600/ocdbt.process_0/d/ecad848186d9d90a59f12b6ad4c3af95 +0 -0
- checkpoint_18600/ocdbt.process_0/d/f7f936a58c481f52b80244293d4e90e2 +0 -0
- checkpoint_18600/ocdbt.process_0/d/fcd663198585c77ac6118d11b26d98d1 +0 -0
- checkpoint_18600/ocdbt.process_0/manifest.ocdbt +0 -0
- meta.json +1 -1
checkpoint_18600/_CHECKPOINT_METADATA
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"item_handlers": "orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler", "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1758133455378462902, "commit_timestamp_nsecs": 1758133455661174697, "custom_metadata": {}}
|
checkpoint_18600/_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_18600/_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_18600/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_18600/d/e8fb8e2162221b8b4302e5b519d00d31
ADDED
|
Binary file (27.1 kB). View file
|
|
|
checkpoint_18600/manifest.ocdbt
ADDED
|
Binary file (118 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/0269e8247fae2ea597c436f321b5869e
ADDED
|
Binary file (640 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/0c7c1ae583522307260686d6c3a7c0b7
ADDED
|
Binary file (614 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/1a3f971051c91416d9e759fff379296b
ADDED
|
Binary file (607 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/378945e046b553d80898ea302983eeca
ADDED
|
Binary file (654 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/39a4c57f34c8a68743f73a6791a4d74a
ADDED
|
Binary file (697 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/43139bbffdfc30d4c68eb9d62692f024
ADDED
|
Binary file (650 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/46097bc2711f3f540658dcda7dde68b7
ADDED
|
Binary file (625 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/48589a405f7063fe1c8658c17fbfa08b
ADDED
|
Binary file (171 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/4a43a24736782dc9c8343725bb54c427
ADDED
|
Binary file (452 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/6e0dbd3904372f423da34b39e52374c5
ADDED
|
Binary file (633 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/759214f7539ef1c601912593389dabba
ADDED
|
Binary file (650 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/851d172ed5a8bebf5738800f9b928182
ADDED
|
Binary file (625 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/88a52735f04097dc8c7297712cbebdea
ADDED
|
Binary file (656 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/8f6d622c66c039fbbe7388bd91cd82b9
ADDED
|
Binary file (629 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/93882eade5f9788c7a685306c6189e4c
ADDED
|
Binary file (682 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/9a33d4987f10544ce76c12a17b1cc16d
ADDED
|
Binary file (1.33 kB). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/9d3d87aa350567a46666b19e15adf058
ADDED
|
Binary file (533 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/a599a292c89dc5c53f9af91172b5bd12
ADDED
|
Binary file (681 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/a7ae19bf629bfeb4040f4d8b59573f89
ADDED
|
Binary file (676 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/a89624385b8088f672240bfea3b86939
ADDED
|
Binary file (38.5 kB). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/b0992e85bfe7f52b028c582a4619aa9f
ADDED
|
Binary file (599 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/ba739de292ec6408bbe0dfe5d1148fb5
ADDED
|
Binary file (647 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/c63ceb0c1c72dd9574975e00b15917c2
ADDED
|
Binary file (637 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/c85fd8ff0a933cc6f574b875d0d985c3
ADDED
|
Binary file (667 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/d30e7cfe83ed986597db688d5d6b457d
ADDED
|
Binary file (52.1 kB). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/d39f9542b8b18c235f96e367a8f07085
ADDED
|
Binary file (18.5 kB). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/ea9aef7a542d893d475aae18de959dc0
ADDED
|
Binary file (611 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/eb257f9ab066115b62b3df3f7b2c1c74
ADDED
|
Binary file (614 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/ecad848186d9d90a59f12b6ad4c3af95
ADDED
|
Binary file (657 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/f7f936a58c481f52b80244293d4e90e2
ADDED
|
Binary file (659 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/d/fcd663198585c77ac6118d11b26d98d1
ADDED
|
Binary file (768 Bytes). View file
|
|
|
checkpoint_18600/ocdbt.process_0/manifest.ocdbt
ADDED
|
Binary file (643 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": 0.001,
|
| 10 |
"predict_delta": true,
|
| 11 |
"seasonal_lag": null,
|
| 12 |
"out_mode": "mse",
|