diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..321f63f58b0ebf0d6926ae97adf1d9ccb18ee14e 100644 --- a/.gitattributes +++ b/.gitattributes @@ -33,3 +33,73 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.zip filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text +10000/params/ocdbt.process_0/d/1a0f6bec25de51b3bb04b6ee0b913d24 filter=lfs diff=lfs merge=lfs -text +10000/params/ocdbt.process_0/d/405826d8cfca26f08f124534d54ad173 filter=lfs diff=lfs merge=lfs -text +10000/params/ocdbt.process_0/d/41d28ce317433d63ae2b5622ea10b53b filter=lfs diff=lfs merge=lfs -text +10000/params/ocdbt.process_0/d/6f76cb6d48d39bbc7344a73a6c2d3149 filter=lfs diff=lfs merge=lfs -text +10000/params/ocdbt.process_0/d/9085174bea05943d66ac80652e8cf2ef filter=lfs diff=lfs merge=lfs -text +10000/params/ocdbt.process_0/d/a699328222fc3f44cf5b45043da05070 filter=lfs diff=lfs merge=lfs -text +10000/params/ocdbt.process_0/d/e14516fb27ad096931c5cfc7447cbd14 filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/1307199b4a6eda794d069ced3a31c91a filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/1645e3e1b9cac02b98597f415891200c filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/1f57869536c947fc3560a75f773eb519 filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/295bfbc5c1d9eb8ed7a0f6a1063691a8 filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/3194eb483fb4ade91ddd5763b0e2f64d filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/435c888608bc9e84224e14a36b85d3d8 filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/475b281c9082dca747199a82f1ffd921 filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/4927e3db53420634e11c0ee2ebb15a82 filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/5d509566dd278af9a8f5e49691459d3e filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/6a52d64b2d5c5dfe077af570e6d7997f filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/7876128bab218fe464f4c0360be51006 filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/924631996a8573ecd2e4c443bcdee333 filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/9cb8b148a7973634bff12aa7e8974045 filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/b1ad7a1e88967829d31be91f141111b1 filter=lfs diff=lfs merge=lfs -text +10000/train_state/ocdbt.process_0/d/d38bc1ee216b23694db0d176bad605dc filter=lfs diff=lfs merge=lfs -text +11000/params/ocdbt.process_0/d/03f7e4a2d86199610fac2fc5b732d6e4 filter=lfs diff=lfs merge=lfs -text +11000/params/ocdbt.process_0/d/1021dcc35bbda8db387e5b18d5dcc22c filter=lfs diff=lfs merge=lfs -text +11000/params/ocdbt.process_0/d/1ce171aead5123c5105787ff137b538b filter=lfs diff=lfs merge=lfs -text +11000/params/ocdbt.process_0/d/4e8f09f122ddb0719ac3f071f0d5a8d1 filter=lfs diff=lfs merge=lfs -text +11000/params/ocdbt.process_0/d/920dd47ec57a211eb5e6e99ec9f56643 filter=lfs diff=lfs merge=lfs -text +11000/params/ocdbt.process_0/d/9ab548398bb0f427594e561d45a2c5b3 filter=lfs diff=lfs merge=lfs -text +11000/params/ocdbt.process_0/d/a53899e1dee8f370e0413d78a8fd8563 filter=lfs diff=lfs merge=lfs -text +11000/params/ocdbt.process_0/d/d8d947044d997d05e3156643a2970240 filter=lfs diff=lfs merge=lfs -text +11000/params/ocdbt.process_0/d/eb37b6798c25c79cc0d96daa1e2e9628 filter=lfs diff=lfs merge=lfs -text +11000/params/ocdbt.process_0/d/ec9f25f7a1aee9ed0a43844bbdfaf2b5 filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/0416d3656676b629c8a4e1f124080096 filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/3cef59378dc039f723087d6c3ec6f912 filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/41f0a9d57aa241c21a1ebb9cf9f9e420 filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/74b4bf995abd131a23ef1177946e452b filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/b4d7a4219cdb59f6c079e0ba34f1c713 filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/b9c891cb33801daf637b08ded4956b7a filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/ca3446e6f6b8390b11a9d9a0ef5bcdf8 filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/cbed992293f3ab19b90863d25be01cfe filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/d487373e246ae08bc2e35f37e6f90a9e filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/dcfc4687e540750fa4a074d3da15b7b9 filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/dd17614884167f641243bac7be072cb1 filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/de21d92b21fcd3aecdbd3a106d6e2c21 filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/e316cedc88cde5c4b3c6d39486761a6f filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/eb1ead7eb6708d090491fa43910f6ca8 filter=lfs diff=lfs merge=lfs -text +11000/train_state/ocdbt.process_0/d/efca221c22daed4f1214529c33a43041 filter=lfs diff=lfs merge=lfs -text +5000/params/ocdbt.process_0/d/32df0fee70688bdd2ace647a64a18e8f filter=lfs diff=lfs merge=lfs -text +5000/params/ocdbt.process_0/d/5845fb67e1a7192373e6ca9f4bf6d4ce filter=lfs diff=lfs merge=lfs -text +5000/params/ocdbt.process_0/d/61480e7b5b517efc12786e7fa2ce2c9c filter=lfs diff=lfs merge=lfs -text +5000/params/ocdbt.process_0/d/63af9d7a36959845369d5a45aa17ad79 filter=lfs diff=lfs merge=lfs -text +5000/params/ocdbt.process_0/d/aafa75fec409c41d9dca6ba33f0b6baa filter=lfs diff=lfs merge=lfs -text +5000/params/ocdbt.process_0/d/c9e38940c1863fa768c2684ff7c0c09e filter=lfs diff=lfs merge=lfs -text +5000/params/ocdbt.process_0/d/d6b85ffe71d6c1ae0f70155f5248eb05 filter=lfs diff=lfs merge=lfs -text +5000/params/ocdbt.process_0/d/fc11f1f051eac1bbf8cd0c6a03a62a41 filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/0f5181f9ff451b8a8f8ff9679c8cb5c6 filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/39298a92eda90c6cbcd86e5c6be1bd33 filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/3f132fa253cbe9cdb3fedb6523babd90 filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/415d9126b07405df52be070cb0ac638a filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/5f336b2a983cdcad11aa83cbecc89b61 filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/72e228cb96f9e0070f73b720eda6f9b1 filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/7867b5092afd15dd321c5852192561c2 filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/834d2ee970214790e27b5a5cc2ea94cd filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/84e79316947a97b7fff3f2764b985102 filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/84f9865b68b638ced9fac9a5a883fed2 filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/9b14aaab37cfecc4d31e1e521c555f86 filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/a7cae5a53d124d28c8138c9fd72c47c7 filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/c4e019d978d4f8260f94ab8f9ac38914 filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/d53ca3e9a6142c2135930446284c0a4c filter=lfs diff=lfs merge=lfs -text +5000/train_state/ocdbt.process_0/d/e454b21615471a5f1ecb3c59b80ecd4a filter=lfs diff=lfs merge=lfs -text diff --git a/10000/_CHECKPOINT_METADATA b/10000/_CHECKPOINT_METADATA new file mode 100644 index 0000000000000000000000000000000000000000..bb6c5d01098ba6c21725ed740aa56b56057352f6 --- /dev/null +++ b/10000/_CHECKPOINT_METADATA @@ -0,0 +1 @@ +{"item_handlers": {"assets": "openpi.training.checkpoints.CallbackHandler", "params": "orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler", "train_state": "orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler"}, "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1758565709683866768, "commit_timestamp_nsecs": 1758565733949197418, "custom_metadata": {}} \ No newline at end of file diff --git a/10000/assets/mixture_ee21701b/norm_stats.json b/10000/assets/mixture_ee21701b/norm_stats.json new file mode 100644 index 0000000000000000000000000000000000000000..182e621547e131f9dbf8b93f2bda72091492ccf0 --- /dev/null +++ b/10000/assets/mixture_ee21701b/norm_stats.json @@ -0,0 +1,80 @@ +{ + "norm_stats": { + "state": { + "mean": [ + 0.30839523673057556, + 0.0213240347802639, + 0.07868705689907074, + 0.029590394347906113, + -0.1100846603512764, + 0.09494097530841827, + 0.7105839252471924 + ], + "std": [ + 0.05918131768703461, + 0.09403301030397415, + 0.04597510024905205, + 0.13276399672031403, + 0.16933076083660126, + 0.6479158401489258, + 0.3638055622577667 + ], + "q01": [ + 0.16066169729232788, + -0.15075165648162364, + -0.017359924914687873, + -0.30125126256942747, + -0.6063957036495209, + -1.2721172106742857, + 0.05044820368885994 + ], + "q99": [ + 0.4422062016367912, + 0.25288478746414184, + 0.19990537272021172, + 0.396708230304718, + 0.2530777354717255, + 1.7838668998718266, + 1.0111272367253898 + ] + }, + "actions": { + "mean": [ + -0.3079696595668793, + -0.020693296566605568, + -0.07779263705015182, + -0.02917383424937725, + 0.10981453210115433, + -0.09645254164934158, + 0.5775498747825623 + ], + "std": [ + 0.06250035762786865, + 0.0976448655128479, + 0.05080597847700119, + 0.13805648684501648, + 0.17500579357147217, + 0.6581579446792603, + 0.490630567073822 + ], + "q01": [ + -0.4479673915207386, + -0.25993944272994995, + -0.20701760437339545, + -0.41325928606986995, + -0.26991710929870605, + -1.8871609056472778, + 0.0 + ], + "q99": [ + -0.15105407805740834, + 0.15632440288066862, + 0.02809868823736905, + 0.3105662116765977, + 0.608537551498413, + 1.292649861431122, + 0.9998 + ] + } + } +} \ No newline at end of file diff --git a/10000/params/_METADATA b/10000/params/_METADATA new file mode 100644 index 0000000000000000000000000000000000000000..d7ee65f7a6eb937e261ebd086de3e332fb21e018 --- /dev/null +++ b/10000/params/_METADATA @@ -0,0 +1 @@ +{"tree_metadata": {"('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1076]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 1076]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4304, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 16, 72, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 288]}}, "('params', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('params', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1152, 512]}}, "('params', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 64, 1152]}}, "('params', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [257152, 512]}}, "('params', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 256, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2048, 64]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 2048, 64]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048, 4096]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 16384, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}}, "use_zarr3": false, "store_array_data_equal_to_fill_value": true, "custom_metadata": null} \ No newline at end of file diff --git a/10000/params/_sharding b/10000/params/_sharding new file mode 100644 index 0000000000000000000000000000000000000000..2df2a383e2412bac6c7c432773d4453584175688 --- /dev/null +++ b/10000/params/_sharding @@ -0,0 +1 @@ +{"cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcucG9zX2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZW1iZWRkZXIuaW5wdXRfZW1iZWRkaW5nLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZmluYWxfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5nYXRpbmdfZWluc3VtLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5saW5lYXIudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4uYXR0bl92ZWNfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ua3ZfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ucV9laW5zdW0udy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9hdHRlbnRpb25fbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9mZndfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}"} \ No newline at end of file diff --git a/10000/params/array_metadatas/process_0 b/10000/params/array_metadatas/process_0 new file mode 100644 index 0000000000000000000000000000000000000000..0c90735afe75d094cf4bcd644a082a83f3187232 --- /dev/null +++ b/10000/params/array_metadatas/process_0 @@ -0,0 +1 @@ +{"array_metadatas": [{"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 1076], "chunk_shape": [27, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 1152, 1076], "chunk_shape": [27, 1152, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 4304, 288], "chunk_shape": [27, 4304, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 16, 72, 288], "chunk_shape": [27, 16, 72, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 288], "chunk_shape": [14, 14, 3, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.bias.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.kernel.value", "write_shape": [1152, 512], "chunk_shape": [1152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.pos_embedding.value", "write_shape": [1, 64, 1152], "chunk_shape": [1, 64, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [257152, 512], "chunk_shape": [257152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.final_norm.scale.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 8, 256, 512], "chunk_shape": [18, 8, 256, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 2048, 64], "chunk_shape": [18, 2, 1, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 8, 2048, 64], "chunk_shape": [18, 8, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 2048, 4096], "chunk_shape": [18, 2, 2048, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 16384, 512], "chunk_shape": [18, 16384, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}]} \ No newline at end of file diff --git a/10000/params/d/9516cedcb7706a315815b64591588659 b/10000/params/d/9516cedcb7706a315815b64591588659 new file mode 100644 index 0000000000000000000000000000000000000000..cbb03983b645105f875eb2eed9b4799c677c3599 Binary files /dev/null and b/10000/params/d/9516cedcb7706a315815b64591588659 differ diff --git a/10000/params/manifest.ocdbt b/10000/params/manifest.ocdbt new file mode 100644 index 0000000000000000000000000000000000000000..8d1dca7c77c15c1967389b86bd3c8ece7a14829e Binary files /dev/null and b/10000/params/manifest.ocdbt differ diff --git a/10000/params/ocdbt.process_0/d/1a0f6bec25de51b3bb04b6ee0b913d24 b/10000/params/ocdbt.process_0/d/1a0f6bec25de51b3bb04b6ee0b913d24 new file mode 100644 index 0000000000000000000000000000000000000000..612df8da9426ba751e8181d353ebbc12ec2cafa2 --- /dev/null +++ b/10000/params/ocdbt.process_0/d/1a0f6bec25de51b3bb04b6ee0b913d24 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:006571efaae88f5bccebba0800bbf5c062b4cc57d2f90807fdec2c17b56d506a +size 2240632546 diff --git a/10000/params/ocdbt.process_0/d/405826d8cfca26f08f124534d54ad173 b/10000/params/ocdbt.process_0/d/405826d8cfca26f08f124534d54ad173 new file mode 100644 index 0000000000000000000000000000000000000000..771741618c71acf19f3c49f73922caaabde61d92 --- /dev/null +++ b/10000/params/ocdbt.process_0/d/405826d8cfca26f08f124534d54ad173 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:45d35d2415c5a7534b0e4c8a2e2ccf27ec25a8d3b6fe3460310e43d070575a62 +size 1440689274 diff --git a/10000/params/ocdbt.process_0/d/41d28ce317433d63ae2b5622ea10b53b b/10000/params/ocdbt.process_0/d/41d28ce317433d63ae2b5622ea10b53b new file mode 100644 index 0000000000000000000000000000000000000000..f0964d5f68e08c675f673edcdeba61435353c3b4 --- /dev/null +++ b/10000/params/ocdbt.process_0/d/41d28ce317433d63ae2b5622ea10b53b @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:95b10b789b817806be6e7f51196f482ea39075dc114fc1a0619238c0db59e155 +size 668107705 diff --git a/10000/params/ocdbt.process_0/d/6942ae5056c803718ea34aa397c7b32a b/10000/params/ocdbt.process_0/d/6942ae5056c803718ea34aa397c7b32a new file mode 100644 index 0000000000000000000000000000000000000000..b1032967f80ba42d38383bd33f354f993a957e70 Binary files /dev/null and b/10000/params/ocdbt.process_0/d/6942ae5056c803718ea34aa397c7b32a differ diff --git a/10000/params/ocdbt.process_0/d/6f76cb6d48d39bbc7344a73a6c2d3149 b/10000/params/ocdbt.process_0/d/6f76cb6d48d39bbc7344a73a6c2d3149 new file mode 100644 index 0000000000000000000000000000000000000000..de59bf69c5f2066221186064df73e1d54822b484 --- /dev/null +++ b/10000/params/ocdbt.process_0/d/6f76cb6d48d39bbc7344a73a6c2d3149 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a6be4df474a9fc76975782dd1e6c5919c174c133b70de4397fb0b2b262893ff0 +size 2517771459 diff --git a/10000/params/ocdbt.process_0/d/75b99f17a35866ffd90c18bd8b01a1f6 b/10000/params/ocdbt.process_0/d/75b99f17a35866ffd90c18bd8b01a1f6 new file mode 100644 index 0000000000000000000000000000000000000000..a92e4b6e27be8c1f17574253d0e700cff6ca28ed Binary files /dev/null and b/10000/params/ocdbt.process_0/d/75b99f17a35866ffd90c18bd8b01a1f6 differ diff --git a/10000/params/ocdbt.process_0/d/9085174bea05943d66ac80652e8cf2ef b/10000/params/ocdbt.process_0/d/9085174bea05943d66ac80652e8cf2ef new file mode 100644 index 0000000000000000000000000000000000000000..14d8564fdb555173a9957ab401b887c90a5d8693 --- /dev/null +++ b/10000/params/ocdbt.process_0/d/9085174bea05943d66ac80652e8cf2ef @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4391e9cb87e6a98de8c598243a9de70167a1f77dd523ec281594024837f66b97 +size 1680752396 diff --git a/10000/params/ocdbt.process_0/d/a0aaecce30e253011fe44b575b63d036 b/10000/params/ocdbt.process_0/d/a0aaecce30e253011fe44b575b63d036 new file mode 100644 index 0000000000000000000000000000000000000000..15aaae2455bbb76aad4af551b7ecedd4c25aaf6b Binary files /dev/null and b/10000/params/ocdbt.process_0/d/a0aaecce30e253011fe44b575b63d036 differ diff --git a/10000/params/ocdbt.process_0/d/a699328222fc3f44cf5b45043da05070 b/10000/params/ocdbt.process_0/d/a699328222fc3f44cf5b45043da05070 new file mode 100644 index 0000000000000000000000000000000000000000..c1b3e896e98086c95524ec978f8413b70408ae8e --- /dev/null +++ b/10000/params/ocdbt.process_0/d/a699328222fc3f44cf5b45043da05070 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1bc09b462a441cb6737a0671f44c0cfa09782885f2126aef93701ca9d6aa84c9 +size 61907165 diff --git a/10000/params/ocdbt.process_0/d/b6724c026fc04342e0538cc2671fac33 b/10000/params/ocdbt.process_0/d/b6724c026fc04342e0538cc2671fac33 new file mode 100644 index 0000000000000000000000000000000000000000..00b7c78f54335798b096323613979880e75a43cd Binary files /dev/null and b/10000/params/ocdbt.process_0/d/b6724c026fc04342e0538cc2671fac33 differ diff --git a/10000/params/ocdbt.process_0/d/e14516fb27ad096931c5cfc7447cbd14 b/10000/params/ocdbt.process_0/d/e14516fb27ad096931c5cfc7447cbd14 new file mode 100644 index 0000000000000000000000000000000000000000..bc703c166abf86f9307d8a1d825a5f8867ce7a34 --- /dev/null +++ b/10000/params/ocdbt.process_0/d/e14516fb27ad096931c5cfc7447cbd14 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1e547869d6b9e5e84a90996d06d72c8e67b940cb7ac05bdeddcdbe0fcdbbd32 +size 2240636184 diff --git a/10000/params/ocdbt.process_0/manifest.ocdbt b/10000/params/ocdbt.process_0/manifest.ocdbt new file mode 100644 index 0000000000000000000000000000000000000000..e9affd9b3b042b4e8e8f7b5ee4a9fa9ac2764414 Binary files /dev/null and b/10000/params/ocdbt.process_0/manifest.ocdbt differ diff --git a/10000/train_state/_METADATA b/10000/train_state/_METADATA new file mode 100644 index 0000000000000000000000000000000000000000..ec9c7af7a310de1fdbd88a0fb82a39b7f5dd4af9 --- /dev/null +++ b/10000/train_state/_METADATA @@ -0,0 +1 @@ +{"tree_metadata": {"('step',)": {"key_metadata": [{"key": "step", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": []}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1076]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 1076]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4304, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 16, 72, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 288]}}, "('params', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('params', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1152, 512]}}, "('params', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 64, 1152]}}, "('params', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [257152, 512]}}, "('params', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 256, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2048, 64]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 2048, 64]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048, 4096]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 16384, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('model_def',)": {"key_metadata": [{"key": "model_def", "key_type": 2}], "value_metadata": {"value_type": "None", "skip_deserialize": true}}, "('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', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1076]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 1076]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4304, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 16, 72, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1152, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 64, 1152]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [257152, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 256, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2048, 64]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 2048, 64]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048, 4096]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 16384, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1076]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 1076]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4304, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 16, 72, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1152, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 64, 1152]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [257152, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 256, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2048, 64]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 2048, 64]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048, 4096]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 16384, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('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', 'count')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "2", "key_type": 1}, {"key": "count", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": []}}, "('ema_params',)": {"key_metadata": [{"key": "ema_params", "key_type": 2}], "value_metadata": {"value_type": "None", "skip_deserialize": true}}}, "use_zarr3": false, "store_array_data_equal_to_fill_value": true, "custom_metadata": null} \ No newline at end of file diff --git a/10000/train_state/_sharding b/10000/train_state/_sharding new file mode 100644 index 0000000000000000000000000000000000000000..a7a7ba4fbfbe4420b2930de10b935d2c467ee425 --- /dev/null +++ b/10000/train_state/_sharding @@ -0,0 +1 @@ +{"b3B0X3N0YXRlLjEuMC5jb3VudA==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJfbm9ybS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMC5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzAuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzAua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzEuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzEua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAua2V5LmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAua2V5Lmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAub3V0LmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAub3V0Lmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAucXVlcnkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAucXVlcnkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAudmFsdWUuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAudmFsdWUua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLmVtYmVkZGluZy5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLmVtYmVkZGluZy5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLmhlYWQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLmhlYWQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLnBvc19lbWJlZGRpbmcudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmVtYmVkZGVyLmlucHV0X2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmZpbmFsX25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLmF0dG5fdmVjX2VpbnN1bS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLmt2X2VpbnN1bS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLnFfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5tbHAuZ2F0aW5nX2VpbnN1bS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5tbHAubGluZWFyLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5wcmVfYXR0ZW50aW9uX25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5wcmVfZmZ3X25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJfbm9ybS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMC5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzAuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzAua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzEuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzEua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAua2V5LmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAua2V5Lmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAub3V0LmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAub3V0Lmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAucXVlcnkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAucXVlcnkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAudmFsdWUuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAudmFsdWUua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLmVtYmVkZGluZy5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLmVtYmVkZGluZy5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLmhlYWQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLmhlYWQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLnBvc19lbWJlZGRpbmcudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmVtYmVkZGVyLmlucHV0X2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmZpbmFsX25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLmF0dG5fdmVjX2VpbnN1bS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLmt2X2VpbnN1bS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLnFfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5tbHAuZ2F0aW5nX2VpbnN1bS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5tbHAubGluZWFyLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5wcmVfYXR0ZW50aW9uX25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5wcmVfZmZ3X25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMi5jb3VudA==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","c3RlcA==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcucG9zX2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZW1iZWRkZXIuaW5wdXRfZW1iZWRkaW5nLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZmluYWxfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5nYXRpbmdfZWluc3VtLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5saW5lYXIudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4uYXR0bl92ZWNfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ua3ZfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ucV9laW5zdW0udy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9hdHRlbnRpb25fbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9mZndfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}"} \ No newline at end of file diff --git a/10000/train_state/array_metadatas/process_0 b/10000/train_state/array_metadatas/process_0 new file mode 100644 index 0000000000000000000000000000000000000000..9a1f979d0388a8b0b34ba00b6e06c02a0127ff69 --- /dev/null +++ b/10000/train_state/array_metadatas/process_0 @@ -0,0 +1 @@ +{"array_metadatas": [{"array_metadata": {"param_name": "step", "write_shape": [], "chunk_shape": [], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 1076], "chunk_shape": [27, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 1152, 1076], "chunk_shape": [27, 1152, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 4304, 288], "chunk_shape": [27, 4304, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 16, 72, 288], "chunk_shape": [27, 16, 72, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 288], "chunk_shape": [14, 14, 3, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.bias.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.kernel.value", "write_shape": [1152, 512], "chunk_shape": [1152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.pos_embedding.value", "write_shape": [1, 64, 1152], "chunk_shape": [1, 64, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [257152, 512], "chunk_shape": [257152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.final_norm.scale.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 8, 256, 512], "chunk_shape": [18, 8, 256, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 2048, 64], "chunk_shape": [18, 2, 1, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 8, 2048, 64], "chunk_shape": [18, 8, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 2048, 4096], "chunk_shape": [18, 2, 2048, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 16384, 512], "chunk_shape": [18, 16384, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "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.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 1076], "chunk_shape": [27, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 1152, 1076], "chunk_shape": [27, 1152, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 4304, 288], "chunk_shape": [27, 4304, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 16, 72, 288], "chunk_shape": [27, 16, 72, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.embedding.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 288], "chunk_shape": [14, 14, 3, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.head.bias.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.head.kernel.value", "write_shape": [1152, 512], "chunk_shape": [1152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.pos_embedding.value", "write_shape": [1, 64, 1152], "chunk_shape": [1, 64, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [257152, 512], "chunk_shape": [257152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.final_norm.scale.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 8, 256, 512], "chunk_shape": [18, 8, 256, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 2048, 64], "chunk_shape": [18, 2, 1, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 8, 2048, 64], "chunk_shape": [18, 8, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 2048, 4096], "chunk_shape": [18, 2, 2048, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 16384, 512], "chunk_shape": [18, 16384, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 1076], "chunk_shape": [27, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 1152, 1076], "chunk_shape": [27, 1152, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 4304, 288], "chunk_shape": [27, 4304, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 16, 72, 288], "chunk_shape": [27, 16, 72, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.embedding.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 288], "chunk_shape": [14, 14, 3, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.head.bias.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.head.kernel.value", "write_shape": [1152, 512], "chunk_shape": [1152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.pos_embedding.value", "write_shape": [1, 64, 1152], "chunk_shape": [1, 64, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [257152, 512], "chunk_shape": [257152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.final_norm.scale.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 8, 256, 512], "chunk_shape": [18, 8, 256, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 2048, 64], "chunk_shape": [18, 2, 1, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 8, 2048, 64], "chunk_shape": [18, 8, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 2048, 4096], "chunk_shape": [18, 2, 2048, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 16384, 512], "chunk_shape": [18, 16384, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.2.count", "write_shape": [], "chunk_shape": [], "ext_metadata": null}}]} \ No newline at end of file diff --git a/10000/train_state/d/1ca67b20d76616fb86bf577f4de0a34e b/10000/train_state/d/1ca67b20d76616fb86bf577f4de0a34e new file mode 100644 index 0000000000000000000000000000000000000000..3ac4f1aab40e90c284776e2f76a41be6f4ad238c Binary files /dev/null and b/10000/train_state/d/1ca67b20d76616fb86bf577f4de0a34e differ diff --git a/10000/train_state/manifest.ocdbt b/10000/train_state/manifest.ocdbt new file mode 100644 index 0000000000000000000000000000000000000000..36b846553090795809ece4d58ff12c87b4e3f944 Binary files /dev/null and b/10000/train_state/manifest.ocdbt differ diff --git a/10000/train_state/ocdbt.process_0/d/1307199b4a6eda794d069ced3a31c91a b/10000/train_state/ocdbt.process_0/d/1307199b4a6eda794d069ced3a31c91a new file mode 100644 index 0000000000000000000000000000000000000000..702a4dc59e5190e9bc95aacff54fe54b89b140ae --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/1307199b4a6eda794d069ced3a31c91a @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4cf072ed5a5148381d5422f335da41e95974ab73be731349e09e3d85f607c1b1 +size 1943904272 diff --git a/10000/train_state/ocdbt.process_0/d/1645e3e1b9cac02b98597f415891200c b/10000/train_state/ocdbt.process_0/d/1645e3e1b9cac02b98597f415891200c new file mode 100644 index 0000000000000000000000000000000000000000..7053242bd1dbe790772a3256d73387b8b51903c9 --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/1645e3e1b9cac02b98597f415891200c @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:79bd646141fcf394f03c3eb0be1c7e30c2c4f4682d2a9f41e624112fff9587f1 +size 2213879410 diff --git a/10000/train_state/ocdbt.process_0/d/1f57869536c947fc3560a75f773eb519 b/10000/train_state/ocdbt.process_0/d/1f57869536c947fc3560a75f773eb519 new file mode 100644 index 0000000000000000000000000000000000000000..362b8548355418deb6dd1376f58aa25687eb118b --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/1f57869536c947fc3560a75f773eb519 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68b7b5cd8e366259f269dc77f98566ec63eacf286b516f503f259c8b490af6f1 +size 2179683566 diff --git a/10000/train_state/ocdbt.process_0/d/295bfbc5c1d9eb8ed7a0f6a1063691a8 b/10000/train_state/ocdbt.process_0/d/295bfbc5c1d9eb8ed7a0f6a1063691a8 new file mode 100644 index 0000000000000000000000000000000000000000..9f3d7cb93cc973edd1f1a009761b6d17683d2fbe --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/295bfbc5c1d9eb8ed7a0f6a1063691a8 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68aad349373b16ff68d208300980d45e91bb2b5b7ac6d1f8657ef07e33d9e765 +size 2582712428 diff --git a/10000/train_state/ocdbt.process_0/d/3194eb483fb4ade91ddd5763b0e2f64d b/10000/train_state/ocdbt.process_0/d/3194eb483fb4ade91ddd5763b0e2f64d new file mode 100644 index 0000000000000000000000000000000000000000..af84c6546d981c07842d5cd77e7c0ba97f0e11b8 --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/3194eb483fb4ade91ddd5763b0e2f64d @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:463d58b9613216353cb55fbf086c121102d37a13e4c4b0dc6a35d8818304f552 +size 2211225899 diff --git a/10000/train_state/ocdbt.process_0/d/435c888608bc9e84224e14a36b85d3d8 b/10000/train_state/ocdbt.process_0/d/435c888608bc9e84224e14a36b85d3d8 new file mode 100644 index 0000000000000000000000000000000000000000..d38aa4e627d5b9d2a94f0fd6247c71c8400c533f --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/435c888608bc9e84224e14a36b85d3d8 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:75e8615a0bf2861a7dc8887d7bbc093978009bb1777e491b3e470be68db5c678 +size 2243393848 diff --git a/10000/train_state/ocdbt.process_0/d/475b281c9082dca747199a82f1ffd921 b/10000/train_state/ocdbt.process_0/d/475b281c9082dca747199a82f1ffd921 new file mode 100644 index 0000000000000000000000000000000000000000..fdbcf1c3d458ea4d519d82210d8a329e07f1b770 --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/475b281c9082dca747199a82f1ffd921 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:afd4454d5bf80bae0e11ae00453a2c503831b736efd3e8dd5f72361ddaded7bd +size 2234335259 diff --git a/10000/train_state/ocdbt.process_0/d/4927e3db53420634e11c0ee2ebb15a82 b/10000/train_state/ocdbt.process_0/d/4927e3db53420634e11c0ee2ebb15a82 new file mode 100644 index 0000000000000000000000000000000000000000..9caf00913d12e3454e046977b401b7d2049318cf --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/4927e3db53420634e11c0ee2ebb15a82 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7c97c7ee81e766476df4d62b868384564107ed3404d9a3415ea0e445e4da2be1 +size 2533672719 diff --git a/10000/train_state/ocdbt.process_0/d/4a65c89c6e8e07e6321eb5755092be4b b/10000/train_state/ocdbt.process_0/d/4a65c89c6e8e07e6321eb5755092be4b new file mode 100644 index 0000000000000000000000000000000000000000..cf91f13a0b34fc6c7dc9244f05d1e15ed166c320 Binary files /dev/null and b/10000/train_state/ocdbt.process_0/d/4a65c89c6e8e07e6321eb5755092be4b differ diff --git a/10000/train_state/ocdbt.process_0/d/58480e959d96ce8f0ca6fb8b96be02a1 b/10000/train_state/ocdbt.process_0/d/58480e959d96ce8f0ca6fb8b96be02a1 new file mode 100644 index 0000000000000000000000000000000000000000..0cc40eebf2f4d30ce258d4e4556e0ea8167aaf28 Binary files /dev/null and b/10000/train_state/ocdbt.process_0/d/58480e959d96ce8f0ca6fb8b96be02a1 differ diff --git a/10000/train_state/ocdbt.process_0/d/5d509566dd278af9a8f5e49691459d3e b/10000/train_state/ocdbt.process_0/d/5d509566dd278af9a8f5e49691459d3e new file mode 100644 index 0000000000000000000000000000000000000000..6875cc6299cbd5198d36e5abac093842518871a4 --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/5d509566dd278af9a8f5e49691459d3e @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:322d3ab189d594428c008fce22600262cf24b350cba6577be6199c33267e659b +size 2226509958 diff --git a/10000/train_state/ocdbt.process_0/d/64e9c5c1e1dc476f0ef2bc30d5c180c3 b/10000/train_state/ocdbt.process_0/d/64e9c5c1e1dc476f0ef2bc30d5c180c3 new file mode 100644 index 0000000000000000000000000000000000000000..a9e69151ed6f275a26c0f30c8d41212162963267 Binary files /dev/null and b/10000/train_state/ocdbt.process_0/d/64e9c5c1e1dc476f0ef2bc30d5c180c3 differ diff --git a/10000/train_state/ocdbt.process_0/d/6a52d64b2d5c5dfe077af570e6d7997f b/10000/train_state/ocdbt.process_0/d/6a52d64b2d5c5dfe077af570e6d7997f new file mode 100644 index 0000000000000000000000000000000000000000..392128ee12a7d9dced8ead97c8e1d270607b5cbf --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/6a52d64b2d5c5dfe077af570e6d7997f @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:460af683a4564b034dc09b023654c4a42cc9b403108673fcf63c2ff6a623289b +size 2240647305 diff --git a/10000/train_state/ocdbt.process_0/d/7876128bab218fe464f4c0360be51006 b/10000/train_state/ocdbt.process_0/d/7876128bab218fe464f4c0360be51006 new file mode 100644 index 0000000000000000000000000000000000000000..2af9d05b3df68eef1d51a918e6d789cc70e72404 --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/7876128bab218fe464f4c0360be51006 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa40bd1fc4dd2bca157bfa613e7947043d832c1ba46e0f6e1904de150aa9ad3c +size 4788789 diff --git a/10000/train_state/ocdbt.process_0/d/924631996a8573ecd2e4c443bcdee333 b/10000/train_state/ocdbt.process_0/d/924631996a8573ecd2e4c443bcdee333 new file mode 100644 index 0000000000000000000000000000000000000000..13b9ceea4e3013dfc237c3ab0e4d0fef289ec6dc --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/924631996a8573ecd2e4c443bcdee333 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1cb09504c2d939059e4ab0dc095479eee71b7d9d3053ab86b39cda23536ee5a5 +size 2246218395 diff --git a/10000/train_state/ocdbt.process_0/d/9cb8b148a7973634bff12aa7e8974045 b/10000/train_state/ocdbt.process_0/d/9cb8b148a7973634bff12aa7e8974045 new file mode 100644 index 0000000000000000000000000000000000000000..069d99307ab27301cca2781ccb231eb6a6fda2aa --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/9cb8b148a7973634bff12aa7e8974045 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0c77bd8bbc25e154ad6c12993f47a6fbe0484015c00759ac1a0abd8cd4cdaf54 +size 2499423326 diff --git a/10000/train_state/ocdbt.process_0/d/b1ad7a1e88967829d31be91f141111b1 b/10000/train_state/ocdbt.process_0/d/b1ad7a1e88967829d31be91f141111b1 new file mode 100644 index 0000000000000000000000000000000000000000..5f060745998b12b407a9ea94d42c9eddf37ed125 --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/b1ad7a1e88967829d31be91f141111b1 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5fa7e0e251a27c04e1ff835776bd4b5b76885917feec84120103dd8f6ce070bf +size 2181537595 diff --git a/10000/train_state/ocdbt.process_0/d/d38bc1ee216b23694db0d176bad605dc b/10000/train_state/ocdbt.process_0/d/d38bc1ee216b23694db0d176bad605dc new file mode 100644 index 0000000000000000000000000000000000000000..c254b66738a446c6cd268bbb84a31a8f20611890 --- /dev/null +++ b/10000/train_state/ocdbt.process_0/d/d38bc1ee216b23694db0d176bad605dc @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:00f8978fa756302efe945ad453e607f5bc74cbbb557446d1ecc7272e617dd888 +size 2592161336 diff --git a/10000/train_state/ocdbt.process_0/d/e666845ac673891b10c1faa01fb3ba19 b/10000/train_state/ocdbt.process_0/d/e666845ac673891b10c1faa01fb3ba19 new file mode 100644 index 0000000000000000000000000000000000000000..6953c03d90be3b5cc2edc8fc02ecd5b57e67e1e3 Binary files /dev/null and b/10000/train_state/ocdbt.process_0/d/e666845ac673891b10c1faa01fb3ba19 differ diff --git a/10000/train_state/ocdbt.process_0/d/ea6147400ef34fde249aa8f43fb1a6db b/10000/train_state/ocdbt.process_0/d/ea6147400ef34fde249aa8f43fb1a6db new file mode 100644 index 0000000000000000000000000000000000000000..a9d25e5229b238fa940432b7701b240bbc60c289 Binary files /dev/null and b/10000/train_state/ocdbt.process_0/d/ea6147400ef34fde249aa8f43fb1a6db differ diff --git a/10000/train_state/ocdbt.process_0/manifest.ocdbt b/10000/train_state/ocdbt.process_0/manifest.ocdbt new file mode 100644 index 0000000000000000000000000000000000000000..023ebf5f1d0acb1a1d80aa17bfc7babf51513ccb Binary files /dev/null and b/10000/train_state/ocdbt.process_0/manifest.ocdbt differ diff --git a/11000/_CHECKPOINT_METADATA b/11000/_CHECKPOINT_METADATA new file mode 100644 index 0000000000000000000000000000000000000000..74878945eab899ce561957934cb84494706df783 --- /dev/null +++ b/11000/_CHECKPOINT_METADATA @@ -0,0 +1 @@ +{"item_handlers": {"assets": "openpi.training.checkpoints.CallbackHandler", "params": "orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler", "train_state": "orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler"}, "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1758567522197132820, "commit_timestamp_nsecs": 1758567548719092821, "custom_metadata": {}} \ No newline at end of file diff --git a/11000/assets/mixture_ee21701b/norm_stats.json b/11000/assets/mixture_ee21701b/norm_stats.json new file mode 100644 index 0000000000000000000000000000000000000000..182e621547e131f9dbf8b93f2bda72091492ccf0 --- /dev/null +++ b/11000/assets/mixture_ee21701b/norm_stats.json @@ -0,0 +1,80 @@ +{ + "norm_stats": { + "state": { + "mean": [ + 0.30839523673057556, + 0.0213240347802639, + 0.07868705689907074, + 0.029590394347906113, + -0.1100846603512764, + 0.09494097530841827, + 0.7105839252471924 + ], + "std": [ + 0.05918131768703461, + 0.09403301030397415, + 0.04597510024905205, + 0.13276399672031403, + 0.16933076083660126, + 0.6479158401489258, + 0.3638055622577667 + ], + "q01": [ + 0.16066169729232788, + -0.15075165648162364, + -0.017359924914687873, + -0.30125126256942747, + -0.6063957036495209, + -1.2721172106742857, + 0.05044820368885994 + ], + "q99": [ + 0.4422062016367912, + 0.25288478746414184, + 0.19990537272021172, + 0.396708230304718, + 0.2530777354717255, + 1.7838668998718266, + 1.0111272367253898 + ] + }, + "actions": { + "mean": [ + -0.3079696595668793, + -0.020693296566605568, + -0.07779263705015182, + -0.02917383424937725, + 0.10981453210115433, + -0.09645254164934158, + 0.5775498747825623 + ], + "std": [ + 0.06250035762786865, + 0.0976448655128479, + 0.05080597847700119, + 0.13805648684501648, + 0.17500579357147217, + 0.6581579446792603, + 0.490630567073822 + ], + "q01": [ + -0.4479673915207386, + -0.25993944272994995, + -0.20701760437339545, + -0.41325928606986995, + -0.26991710929870605, + -1.8871609056472778, + 0.0 + ], + "q99": [ + -0.15105407805740834, + 0.15632440288066862, + 0.02809868823736905, + 0.3105662116765977, + 0.608537551498413, + 1.292649861431122, + 0.9998 + ] + } + } +} \ No newline at end of file diff --git a/11000/params/_METADATA b/11000/params/_METADATA new file mode 100644 index 0000000000000000000000000000000000000000..d7ee65f7a6eb937e261ebd086de3e332fb21e018 --- /dev/null +++ b/11000/params/_METADATA @@ -0,0 +1 @@ +{"tree_metadata": {"('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1076]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 1076]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4304, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 16, 72, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 288]}}, "('params', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('params', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1152, 512]}}, "('params', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 64, 1152]}}, "('params', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [257152, 512]}}, "('params', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 256, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2048, 64]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 2048, 64]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048, 4096]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 16384, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}}, "use_zarr3": false, "store_array_data_equal_to_fill_value": true, "custom_metadata": null} \ No newline at end of file diff --git a/11000/params/_sharding b/11000/params/_sharding new file mode 100644 index 0000000000000000000000000000000000000000..2df2a383e2412bac6c7c432773d4453584175688 --- /dev/null +++ b/11000/params/_sharding @@ -0,0 +1 @@ +{"cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcucG9zX2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZW1iZWRkZXIuaW5wdXRfZW1iZWRkaW5nLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZmluYWxfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5nYXRpbmdfZWluc3VtLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5saW5lYXIudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4uYXR0bl92ZWNfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ua3ZfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ucV9laW5zdW0udy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9hdHRlbnRpb25fbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9mZndfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}"} \ No newline at end of file diff --git a/11000/params/array_metadatas/process_0 b/11000/params/array_metadatas/process_0 new file mode 100644 index 0000000000000000000000000000000000000000..0c90735afe75d094cf4bcd644a082a83f3187232 --- /dev/null +++ b/11000/params/array_metadatas/process_0 @@ -0,0 +1 @@ +{"array_metadatas": [{"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 1076], "chunk_shape": [27, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 1152, 1076], "chunk_shape": [27, 1152, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 4304, 288], "chunk_shape": [27, 4304, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 16, 72, 288], "chunk_shape": [27, 16, 72, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 288], "chunk_shape": [14, 14, 3, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.bias.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.kernel.value", "write_shape": [1152, 512], "chunk_shape": [1152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.pos_embedding.value", "write_shape": [1, 64, 1152], "chunk_shape": [1, 64, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [257152, 512], "chunk_shape": [257152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.final_norm.scale.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 8, 256, 512], "chunk_shape": [18, 8, 256, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 2048, 64], "chunk_shape": [18, 2, 1, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 8, 2048, 64], "chunk_shape": [18, 8, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 2048, 4096], "chunk_shape": [18, 2, 2048, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 16384, 512], "chunk_shape": [18, 16384, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}]} \ No newline at end of file diff --git a/11000/params/d/3d8869df8ec9b4b43001b2af0857f5b7 b/11000/params/d/3d8869df8ec9b4b43001b2af0857f5b7 new file mode 100644 index 0000000000000000000000000000000000000000..e68b27744613be83a59deaf8563f354b3615ca66 Binary files /dev/null and b/11000/params/d/3d8869df8ec9b4b43001b2af0857f5b7 differ diff --git a/11000/params/manifest.ocdbt b/11000/params/manifest.ocdbt new file mode 100644 index 0000000000000000000000000000000000000000..5802d1eb31560d9f0b3ad5fa8a89537f03a4669f Binary files /dev/null and b/11000/params/manifest.ocdbt differ diff --git a/11000/params/ocdbt.process_0/d/03f7e4a2d86199610fac2fc5b732d6e4 b/11000/params/ocdbt.process_0/d/03f7e4a2d86199610fac2fc5b732d6e4 new file mode 100644 index 0000000000000000000000000000000000000000..7fd4f781dd4d4a5682cd2eeb19e5dcfbf1d866ed --- /dev/null +++ b/11000/params/ocdbt.process_0/d/03f7e4a2d86199610fac2fc5b732d6e4 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5ecc9c6ae81141536820f6ecaa8f6eb5af66b4c981606c21d17979bef4670eb9 +size 48632151 diff --git a/11000/params/ocdbt.process_0/d/09d94a208bd5bd9ac53f28ba43d1429c b/11000/params/ocdbt.process_0/d/09d94a208bd5bd9ac53f28ba43d1429c new file mode 100644 index 0000000000000000000000000000000000000000..da58ef8951326723b242fa6ccc4969f4c9b1cc8d Binary files /dev/null and b/11000/params/ocdbt.process_0/d/09d94a208bd5bd9ac53f28ba43d1429c differ diff --git a/11000/params/ocdbt.process_0/d/1021dcc35bbda8db387e5b18d5dcc22c b/11000/params/ocdbt.process_0/d/1021dcc35bbda8db387e5b18d5dcc22c new file mode 100644 index 0000000000000000000000000000000000000000..acff45df9655e54aa703243b9ac2f9c16076793f --- /dev/null +++ b/11000/params/ocdbt.process_0/d/1021dcc35bbda8db387e5b18d5dcc22c @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dba02ba89cfc38bbff0f54fb435a969f9128ad9872888f2a3a004b24b342e156 +size 1120314686 diff --git a/11000/params/ocdbt.process_0/d/1ce171aead5123c5105787ff137b538b b/11000/params/ocdbt.process_0/d/1ce171aead5123c5105787ff137b538b new file mode 100644 index 0000000000000000000000000000000000000000..6a585170a6c2670e61205af78eba55d8b879366a --- /dev/null +++ b/11000/params/ocdbt.process_0/d/1ce171aead5123c5105787ff137b538b @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:336b31d2f33de0b5ec2240d93f6a60a86cdc690cc6833fdafdf90901dce151fd +size 1120327372 diff --git a/11000/params/ocdbt.process_0/d/4e8f09f122ddb0719ac3f071f0d5a8d1 b/11000/params/ocdbt.process_0/d/4e8f09f122ddb0719ac3f071f0d5a8d1 new file mode 100644 index 0000000000000000000000000000000000000000..a0cbcf68d6018b03ddf2fef7ee00aba798eb2363 --- /dev/null +++ b/11000/params/ocdbt.process_0/d/4e8f09f122ddb0719ac3f071f0d5a8d1 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8fceb3f5f549bdeb7ae9055b0f00245e7f913a8cb7e3a4553ecccfaaee35b502 +size 560270829 diff --git a/11000/params/ocdbt.process_0/d/8f0922d42b9211e713da2426c42db80e b/11000/params/ocdbt.process_0/d/8f0922d42b9211e713da2426c42db80e new file mode 100644 index 0000000000000000000000000000000000000000..b1032967f80ba42d38383bd33f354f993a957e70 Binary files /dev/null and b/11000/params/ocdbt.process_0/d/8f0922d42b9211e713da2426c42db80e differ diff --git a/11000/params/ocdbt.process_0/d/920dd47ec57a211eb5e6e99ec9f56643 b/11000/params/ocdbt.process_0/d/920dd47ec57a211eb5e6e99ec9f56643 new file mode 100644 index 0000000000000000000000000000000000000000..a87d8475cb2196e5a13b4c072ee7a4b77a37c4d3 --- /dev/null +++ b/11000/params/ocdbt.process_0/d/920dd47ec57a211eb5e6e99ec9f56643 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5be863009676fa4180ee116dcd1ba710b51d214d8a56f7fef07eb3ec848da0bc +size 301307345 diff --git a/11000/params/ocdbt.process_0/d/9ab548398bb0f427594e561d45a2c5b3 b/11000/params/ocdbt.process_0/d/9ab548398bb0f427594e561d45a2c5b3 new file mode 100644 index 0000000000000000000000000000000000000000..74620570d20aba7bf08697448f62b89835ef0543 --- /dev/null +++ b/11000/params/ocdbt.process_0/d/9ab548398bb0f427594e561d45a2c5b3 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:75656db713c713f7023b747a910256c3be9c7aa2fd5233f850bd8e74c32c7cc8 +size 267252 diff --git a/11000/params/ocdbt.process_0/d/a53899e1dee8f370e0413d78a8fd8563 b/11000/params/ocdbt.process_0/d/a53899e1dee8f370e0413d78a8fd8563 new file mode 100644 index 0000000000000000000000000000000000000000..cbf523661367c9ef4fc1dcfa80ffa17244c22fab --- /dev/null +++ b/11000/params/ocdbt.process_0/d/a53899e1dee8f370e0413d78a8fd8563 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e31a981234fd40b3aa3868871808d77d05102dc546b8a23592d3db81b55245bf +size 2240636894 diff --git a/11000/params/ocdbt.process_0/d/d8d947044d997d05e3156643a2970240 b/11000/params/ocdbt.process_0/d/d8d947044d997d05e3156643a2970240 new file mode 100644 index 0000000000000000000000000000000000000000..c5471190c4fb0ad09f6392587cf1dd4d6599dcfe --- /dev/null +++ b/11000/params/ocdbt.process_0/d/d8d947044d997d05e3156643a2970240 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:795e193c7909f97038f27a2b4f1f8b91a26e6039fbfa842929ebc6676e42500d +size 1723832640 diff --git a/11000/params/ocdbt.process_0/d/eb37b6798c25c79cc0d96daa1e2e9628 b/11000/params/ocdbt.process_0/d/eb37b6798c25c79cc0d96daa1e2e9628 new file mode 100644 index 0000000000000000000000000000000000000000..5c6806394dc46d7aa43bf234469992ef200de6cb --- /dev/null +++ b/11000/params/ocdbt.process_0/d/eb37b6798c25c79cc0d96daa1e2e9628 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8a3113226f5b0086d09c7e985a4113bd38ed894cf6bf359a0d153781bfeaa91e +size 2660086678 diff --git a/11000/params/ocdbt.process_0/d/ec9f25f7a1aee9ed0a43844bbdfaf2b5 b/11000/params/ocdbt.process_0/d/ec9f25f7a1aee9ed0a43844bbdfaf2b5 new file mode 100644 index 0000000000000000000000000000000000000000..3e6510bbabeee24857efb585a2291722203b4ed9 --- /dev/null +++ b/11000/params/ocdbt.process_0/d/ec9f25f7a1aee9ed0a43844bbdfaf2b5 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:89dd0c73a15747ad28ec8329602c2768f4526ec03bc116186b4cc49dd5c8d8db +size 1074856985 diff --git a/11000/params/ocdbt.process_0/manifest.ocdbt b/11000/params/ocdbt.process_0/manifest.ocdbt new file mode 100644 index 0000000000000000000000000000000000000000..60b9f68333fa921f915cd9da0bb23628f2a8d4b9 Binary files /dev/null and b/11000/params/ocdbt.process_0/manifest.ocdbt differ diff --git a/11000/train_state/_METADATA b/11000/train_state/_METADATA new file mode 100644 index 0000000000000000000000000000000000000000..ec9c7af7a310de1fdbd88a0fb82a39b7f5dd4af9 --- /dev/null +++ b/11000/train_state/_METADATA @@ -0,0 +1 @@ +{"tree_metadata": {"('step',)": {"key_metadata": [{"key": "step", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": []}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1076]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 1076]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4304, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 16, 72, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 288]}}, "('params', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('params', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1152, 512]}}, "('params', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 64, 1152]}}, "('params', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [257152, 512]}}, "('params', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 256, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2048, 64]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 2048, 64]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048, 4096]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 16384, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('model_def',)": {"key_metadata": [{"key": "model_def", "key_type": 2}], "value_metadata": {"value_type": "None", "skip_deserialize": true}}, "('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', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1076]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 1076]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4304, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 16, 72, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1152, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 64, 1152]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [257152, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 256, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2048, 64]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 2048, 64]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048, 4096]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 16384, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1076]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 1076]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4304, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 16, 72, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1152, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 64, 1152]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [257152, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 256, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2048, 64]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 2048, 64]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048, 4096]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 16384, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('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', 'count')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "2", "key_type": 1}, {"key": "count", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": []}}, "('ema_params',)": {"key_metadata": [{"key": "ema_params", "key_type": 2}], "value_metadata": {"value_type": "None", "skip_deserialize": true}}}, "use_zarr3": false, "store_array_data_equal_to_fill_value": true, "custom_metadata": null} \ No newline at end of file diff --git a/11000/train_state/_sharding b/11000/train_state/_sharding new file mode 100644 index 0000000000000000000000000000000000000000..a7a7ba4fbfbe4420b2930de10b935d2c467ee425 --- /dev/null +++ b/11000/train_state/_sharding @@ -0,0 +1 @@ +{"b3B0X3N0YXRlLjEuMC5jb3VudA==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJfbm9ybS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMC5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzAuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzAua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzEuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzEua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAua2V5LmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAua2V5Lmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAub3V0LmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAub3V0Lmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAucXVlcnkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAucXVlcnkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAudmFsdWUuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAudmFsdWUua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLmVtYmVkZGluZy5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLmVtYmVkZGluZy5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLmhlYWQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLmhlYWQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLnBvc19lbWJlZGRpbmcudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmVtYmVkZGVyLmlucHV0X2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmZpbmFsX25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLmF0dG5fdmVjX2VpbnN1bS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLmt2X2VpbnN1bS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLnFfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5tbHAuZ2F0aW5nX2VpbnN1bS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5tbHAubGluZWFyLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5wcmVfYXR0ZW50aW9uX25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5wcmVfZmZ3X25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJfbm9ybS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMC5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzAuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzAua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzEuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzEua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAua2V5LmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAua2V5Lmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAub3V0LmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAub3V0Lmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAucXVlcnkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAucXVlcnkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAudmFsdWUuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAudmFsdWUua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLmVtYmVkZGluZy5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLmVtYmVkZGluZy5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLmhlYWQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLmhlYWQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLnBvc19lbWJlZGRpbmcudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmVtYmVkZGVyLmlucHV0X2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmZpbmFsX25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLmF0dG5fdmVjX2VpbnN1bS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLmt2X2VpbnN1bS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLnFfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5tbHAuZ2F0aW5nX2VpbnN1bS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5tbHAubGluZWFyLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5wcmVfYXR0ZW50aW9uX25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5wcmVfZmZ3X25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMi5jb3VudA==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","c3RlcA==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcucG9zX2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZW1iZWRkZXIuaW5wdXRfZW1iZWRkaW5nLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZmluYWxfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5nYXRpbmdfZWluc3VtLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5saW5lYXIudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4uYXR0bl92ZWNfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ua3ZfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ucV9laW5zdW0udy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9hdHRlbnRpb25fbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9mZndfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}"} \ No newline at end of file diff --git a/11000/train_state/array_metadatas/process_0 b/11000/train_state/array_metadatas/process_0 new file mode 100644 index 0000000000000000000000000000000000000000..9a1f979d0388a8b0b34ba00b6e06c02a0127ff69 --- /dev/null +++ b/11000/train_state/array_metadatas/process_0 @@ -0,0 +1 @@ +{"array_metadatas": [{"array_metadata": {"param_name": "step", "write_shape": [], "chunk_shape": [], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 1076], "chunk_shape": [27, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 1152, 1076], "chunk_shape": [27, 1152, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 4304, 288], "chunk_shape": [27, 4304, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 16, 72, 288], "chunk_shape": [27, 16, 72, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 288], "chunk_shape": [14, 14, 3, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.bias.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.kernel.value", "write_shape": [1152, 512], "chunk_shape": [1152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.pos_embedding.value", "write_shape": [1, 64, 1152], "chunk_shape": [1, 64, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [257152, 512], "chunk_shape": [257152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.final_norm.scale.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 8, 256, 512], "chunk_shape": [18, 8, 256, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 2048, 64], "chunk_shape": [18, 2, 1, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 8, 2048, 64], "chunk_shape": [18, 8, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 2048, 4096], "chunk_shape": [18, 2, 2048, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 16384, 512], "chunk_shape": [18, 16384, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "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.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 1076], "chunk_shape": [27, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 1152, 1076], "chunk_shape": [27, 1152, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 4304, 288], "chunk_shape": [27, 4304, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 16, 72, 288], "chunk_shape": [27, 16, 72, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.embedding.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 288], "chunk_shape": [14, 14, 3, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.head.bias.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.head.kernel.value", "write_shape": [1152, 512], "chunk_shape": [1152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.pos_embedding.value", "write_shape": [1, 64, 1152], "chunk_shape": [1, 64, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [257152, 512], "chunk_shape": [257152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.final_norm.scale.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 8, 256, 512], "chunk_shape": [18, 8, 256, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 2048, 64], "chunk_shape": [18, 2, 1, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 8, 2048, 64], "chunk_shape": [18, 8, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 2048, 4096], "chunk_shape": [18, 2, 2048, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 16384, 512], "chunk_shape": [18, 16384, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 1076], "chunk_shape": [27, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 1152, 1076], "chunk_shape": [27, 1152, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 4304, 288], "chunk_shape": [27, 4304, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 16, 72, 288], "chunk_shape": [27, 16, 72, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.embedding.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 288], "chunk_shape": [14, 14, 3, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.head.bias.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.head.kernel.value", "write_shape": [1152, 512], "chunk_shape": [1152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.pos_embedding.value", "write_shape": [1, 64, 1152], "chunk_shape": [1, 64, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [257152, 512], "chunk_shape": [257152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.final_norm.scale.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 8, 256, 512], "chunk_shape": [18, 8, 256, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 2048, 64], "chunk_shape": [18, 2, 1, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 8, 2048, 64], "chunk_shape": [18, 8, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 2048, 4096], "chunk_shape": [18, 2, 2048, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 16384, 512], "chunk_shape": [18, 16384, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.2.count", "write_shape": [], "chunk_shape": [], "ext_metadata": null}}]} \ No newline at end of file diff --git a/11000/train_state/d/bbfd7df6df666b8d134ac312f3dbc8ed b/11000/train_state/d/bbfd7df6df666b8d134ac312f3dbc8ed new file mode 100644 index 0000000000000000000000000000000000000000..d8fdd7e0f3c0320f527289fc6a89f0ebc1eb0e56 Binary files /dev/null and b/11000/train_state/d/bbfd7df6df666b8d134ac312f3dbc8ed differ diff --git a/11000/train_state/manifest.ocdbt b/11000/train_state/manifest.ocdbt new file mode 100644 index 0000000000000000000000000000000000000000..468969e2d8b88b8438647f5020f3632adc8cd500 Binary files /dev/null and b/11000/train_state/manifest.ocdbt differ diff --git a/11000/train_state/ocdbt.process_0/d/0416d3656676b629c8a4e1f124080096 b/11000/train_state/ocdbt.process_0/d/0416d3656676b629c8a4e1f124080096 new file mode 100644 index 0000000000000000000000000000000000000000..8841aeff0bcd8473356e1760621f70d05c658209 --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/0416d3656676b629c8a4e1f124080096 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:919305cd7c051e7ba4b6299e02ea387b52aeb72db238bda7a65efd56227cc64c +size 2243810151 diff --git a/11000/train_state/ocdbt.process_0/d/1715641c2535c8c4bf49978c772c7fdb b/11000/train_state/ocdbt.process_0/d/1715641c2535c8c4bf49978c772c7fdb new file mode 100644 index 0000000000000000000000000000000000000000..77459679cb9f0be577d06a2b4bde74c296cee144 Binary files /dev/null and b/11000/train_state/ocdbt.process_0/d/1715641c2535c8c4bf49978c772c7fdb differ diff --git a/11000/train_state/ocdbt.process_0/d/3cef59378dc039f723087d6c3ec6f912 b/11000/train_state/ocdbt.process_0/d/3cef59378dc039f723087d6c3ec6f912 new file mode 100644 index 0000000000000000000000000000000000000000..bc820d2017cbcc1e2ab0a6aa92f7e3f5f828dd1e --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/3cef59378dc039f723087d6c3ec6f912 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f7b1e1e57d4ecd222208937987727ec76d695540557eafeb364e90603bc367a +size 475437 diff --git a/11000/train_state/ocdbt.process_0/d/41f0a9d57aa241c21a1ebb9cf9f9e420 b/11000/train_state/ocdbt.process_0/d/41f0a9d57aa241c21a1ebb9cf9f9e420 new file mode 100644 index 0000000000000000000000000000000000000000..c688200bbbf082bd394bd6403af6bd03fd19ac6d --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/41f0a9d57aa241c21a1ebb9cf9f9e420 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:307f61d204a01269a87bb3c0b79a6d43d68cd859f3080fdcccdf0159078a68cd +size 2197946382 diff --git a/11000/train_state/ocdbt.process_0/d/6b4f42eab4a35b612d83c8ce6fda00fb b/11000/train_state/ocdbt.process_0/d/6b4f42eab4a35b612d83c8ce6fda00fb new file mode 100644 index 0000000000000000000000000000000000000000..32aeb3b90d60b0a7055cd0d594f6225851c8357d Binary files /dev/null and b/11000/train_state/ocdbt.process_0/d/6b4f42eab4a35b612d83c8ce6fda00fb differ diff --git a/11000/train_state/ocdbt.process_0/d/74b4bf995abd131a23ef1177946e452b b/11000/train_state/ocdbt.process_0/d/74b4bf995abd131a23ef1177946e452b new file mode 100644 index 0000000000000000000000000000000000000000..37a54d9878f8286be6d2c7b7e6544738e75023ce --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/74b4bf995abd131a23ef1177946e452b @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7bf211d5d02529b7eddc88fee64a3f8eb8f1044cf42227ecf2cb8222b6150e42 +size 2571157806 diff --git a/11000/train_state/ocdbt.process_0/d/a7ff46f65569ae1344a859708dd639a3 b/11000/train_state/ocdbt.process_0/d/a7ff46f65569ae1344a859708dd639a3 new file mode 100644 index 0000000000000000000000000000000000000000..9bbb4dec5951505e4d3de5f532dd53f55878c7d3 Binary files /dev/null and b/11000/train_state/ocdbt.process_0/d/a7ff46f65569ae1344a859708dd639a3 differ diff --git a/11000/train_state/ocdbt.process_0/d/b4d7a4219cdb59f6c079e0ba34f1c713 b/11000/train_state/ocdbt.process_0/d/b4d7a4219cdb59f6c079e0ba34f1c713 new file mode 100644 index 0000000000000000000000000000000000000000..a7cd4aeac4cd404c1681d869df9df80bfd473db0 --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/b4d7a4219cdb59f6c079e0ba34f1c713 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:da8aba4671add53f99307cb247e865d4b9deebc70b0277ab56cb8d77e5f36571 +size 2159514350 diff --git a/11000/train_state/ocdbt.process_0/d/b9c891cb33801daf637b08ded4956b7a b/11000/train_state/ocdbt.process_0/d/b9c891cb33801daf637b08ded4956b7a new file mode 100644 index 0000000000000000000000000000000000000000..298057fb998b6af63c6886aee687c684f2cbe25a --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/b9c891cb33801daf637b08ded4956b7a @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64aba7437a5a9cea41f0139bd0e3fa36f74886397867a49c6eba297dabc79628 +size 2179767764 diff --git a/11000/train_state/ocdbt.process_0/d/bf99b8f246967f833d97692bf6ab2614 b/11000/train_state/ocdbt.process_0/d/bf99b8f246967f833d97692bf6ab2614 new file mode 100644 index 0000000000000000000000000000000000000000..a9e69151ed6f275a26c0f30c8d41212162963267 Binary files /dev/null and b/11000/train_state/ocdbt.process_0/d/bf99b8f246967f833d97692bf6ab2614 differ diff --git a/11000/train_state/ocdbt.process_0/d/c7a5f3d7ea2cd40549b56eb78fa96059 b/11000/train_state/ocdbt.process_0/d/c7a5f3d7ea2cd40549b56eb78fa96059 new file mode 100644 index 0000000000000000000000000000000000000000..d0ae45733d3d65aa67ee77b66d740ab7eadec299 Binary files /dev/null and b/11000/train_state/ocdbt.process_0/d/c7a5f3d7ea2cd40549b56eb78fa96059 differ diff --git a/11000/train_state/ocdbt.process_0/d/ca3446e6f6b8390b11a9d9a0ef5bcdf8 b/11000/train_state/ocdbt.process_0/d/ca3446e6f6b8390b11a9d9a0ef5bcdf8 new file mode 100644 index 0000000000000000000000000000000000000000..388bdf0d5ef5316ca2b961107d8cd4c88fcbca94 --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/ca3446e6f6b8390b11a9d9a0ef5bcdf8 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7ae629c5bcd44f72a98b56a693d75f0f6847dee1d816bc84e65e0934a70a625 +size 2240797946 diff --git a/11000/train_state/ocdbt.process_0/d/cbed992293f3ab19b90863d25be01cfe b/11000/train_state/ocdbt.process_0/d/cbed992293f3ab19b90863d25be01cfe new file mode 100644 index 0000000000000000000000000000000000000000..38cbaedae753ffbb070ee17d00613dca805d9a4e --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/cbed992293f3ab19b90863d25be01cfe @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b14f4fe19f761583db908889e94dafe9f2582273545c44c0cfda8e246e184ff +size 2211548655 diff --git a/11000/train_state/ocdbt.process_0/d/d487373e246ae08bc2e35f37e6f90a9e b/11000/train_state/ocdbt.process_0/d/d487373e246ae08bc2e35f37e6f90a9e new file mode 100644 index 0000000000000000000000000000000000000000..cd624cb24716f9d15322070397562be430eeeeb7 --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/d487373e246ae08bc2e35f37e6f90a9e @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6c3b9108a497ca05219ab6f9afdb5bdaf2aee8752a2f132254bc94146f64f788 +size 2179350537 diff --git a/11000/train_state/ocdbt.process_0/d/dcfc4687e540750fa4a074d3da15b7b9 b/11000/train_state/ocdbt.process_0/d/dcfc4687e540750fa4a074d3da15b7b9 new file mode 100644 index 0000000000000000000000000000000000000000..b623ce0bb1a34e93e16d56083819103e61f32289 --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/dcfc4687e540750fa4a074d3da15b7b9 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df1d7f49360d487bf3d77409868c47255c8a741ceea43978e0f82a0030c7f3a5 +size 2709094501 diff --git a/11000/train_state/ocdbt.process_0/d/dd17614884167f641243bac7be072cb1 b/11000/train_state/ocdbt.process_0/d/dd17614884167f641243bac7be072cb1 new file mode 100644 index 0000000000000000000000000000000000000000..6e98d4c2dfcec1148d70dc904baf473b19157545 --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/dd17614884167f641243bac7be072cb1 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c5d7003451ca8804fd980f1e4a784ff98b44caec7146e52c2d6b67172f024b07 +size 2182351112 diff --git a/11000/train_state/ocdbt.process_0/d/de21d92b21fcd3aecdbd3a106d6e2c21 b/11000/train_state/ocdbt.process_0/d/de21d92b21fcd3aecdbd3a106d6e2c21 new file mode 100644 index 0000000000000000000000000000000000000000..b15ddf1871f767877a91cf163531fe2062747fa8 --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/de21d92b21fcd3aecdbd3a106d6e2c21 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c1828e5a067a2a565ab145fb4b90f878f5849bdf402617ded39f753d02fb8170 +size 2177921953 diff --git a/11000/train_state/ocdbt.process_0/d/e316cedc88cde5c4b3c6d39486761a6f b/11000/train_state/ocdbt.process_0/d/e316cedc88cde5c4b3c6d39486761a6f new file mode 100644 index 0000000000000000000000000000000000000000..9aa1db3e4c6849e8f17dab255b4dbe0bc94f3302 --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/e316cedc88cde5c4b3c6d39486761a6f @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68ce626f8123733400f53d4143091a0d1f03ed69da780fa97c35e546327f04f0 +size 2243883724 diff --git a/11000/train_state/ocdbt.process_0/d/eb1ead7eb6708d090491fa43910f6ca8 b/11000/train_state/ocdbt.process_0/d/eb1ead7eb6708d090491fa43910f6ca8 new file mode 100644 index 0000000000000000000000000000000000000000..1f8484ccd1d91a6f45cb948b59821998ee87ae15 --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/eb1ead7eb6708d090491fa43910f6ca8 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37056622c57b87f6d6de5b091e551f6e67d8faa952723e08ba08ac854a65c32b +size 2247020863 diff --git a/11000/train_state/ocdbt.process_0/d/efca221c22daed4f1214529c33a43041 b/11000/train_state/ocdbt.process_0/d/efca221c22daed4f1214529c33a43041 new file mode 100644 index 0000000000000000000000000000000000000000..ad438be58f73d863a82fcad83e12406bcd918072 --- /dev/null +++ b/11000/train_state/ocdbt.process_0/d/efca221c22daed4f1214529c33a43041 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5a433fe921b5188c85f3391a6019134e213c6da3735df2b26f9b8f7d7d6e8f31 +size 2595534893 diff --git a/11000/train_state/ocdbt.process_0/manifest.ocdbt b/11000/train_state/ocdbt.process_0/manifest.ocdbt new file mode 100644 index 0000000000000000000000000000000000000000..42fbf510c2f7b41bdc27b191ed15d40ac090bb5a Binary files /dev/null and b/11000/train_state/ocdbt.process_0/manifest.ocdbt differ diff --git a/5000/_CHECKPOINT_METADATA b/5000/_CHECKPOINT_METADATA new file mode 100644 index 0000000000000000000000000000000000000000..f7cb964ecfda9488ed34e88e8f270805861c98f4 --- /dev/null +++ b/5000/_CHECKPOINT_METADATA @@ -0,0 +1 @@ +{"item_handlers": {"assets": "openpi.training.checkpoints.CallbackHandler", "params": "orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler", "train_state": "orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler"}, "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1758556643381915784, "commit_timestamp_nsecs": 1758556674892908917, "custom_metadata": {}} \ No newline at end of file diff --git a/5000/assets/mixture_ee21701b/norm_stats.json b/5000/assets/mixture_ee21701b/norm_stats.json new file mode 100644 index 0000000000000000000000000000000000000000..182e621547e131f9dbf8b93f2bda72091492ccf0 --- /dev/null +++ b/5000/assets/mixture_ee21701b/norm_stats.json @@ -0,0 +1,80 @@ +{ + "norm_stats": { + "state": { + "mean": [ + 0.30839523673057556, + 0.0213240347802639, + 0.07868705689907074, + 0.029590394347906113, + -0.1100846603512764, + 0.09494097530841827, + 0.7105839252471924 + ], + "std": [ + 0.05918131768703461, + 0.09403301030397415, + 0.04597510024905205, + 0.13276399672031403, + 0.16933076083660126, + 0.6479158401489258, + 0.3638055622577667 + ], + "q01": [ + 0.16066169729232788, + -0.15075165648162364, + -0.017359924914687873, + -0.30125126256942747, + -0.6063957036495209, + -1.2721172106742857, + 0.05044820368885994 + ], + "q99": [ + 0.4422062016367912, + 0.25288478746414184, + 0.19990537272021172, + 0.396708230304718, + 0.2530777354717255, + 1.7838668998718266, + 1.0111272367253898 + ] + }, + "actions": { + "mean": [ + -0.3079696595668793, + -0.020693296566605568, + -0.07779263705015182, + -0.02917383424937725, + 0.10981453210115433, + -0.09645254164934158, + 0.5775498747825623 + ], + "std": [ + 0.06250035762786865, + 0.0976448655128479, + 0.05080597847700119, + 0.13805648684501648, + 0.17500579357147217, + 0.6581579446792603, + 0.490630567073822 + ], + "q01": [ + -0.4479673915207386, + -0.25993944272994995, + -0.20701760437339545, + -0.41325928606986995, + -0.26991710929870605, + -1.8871609056472778, + 0.0 + ], + "q99": [ + -0.15105407805740834, + 0.15632440288066862, + 0.02809868823736905, + 0.3105662116765977, + 0.608537551498413, + 1.292649861431122, + 0.9998 + ] + } + } +} \ No newline at end of file diff --git a/5000/params/_METADATA b/5000/params/_METADATA new file mode 100644 index 0000000000000000000000000000000000000000..d7ee65f7a6eb937e261ebd086de3e332fb21e018 --- /dev/null +++ b/5000/params/_METADATA @@ -0,0 +1 @@ +{"tree_metadata": {"('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1076]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 1076]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4304, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 16, 72, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 288]}}, "('params', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('params', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1152, 512]}}, "('params', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 64, 1152]}}, "('params', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [257152, 512]}}, "('params', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 256, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2048, 64]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 2048, 64]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048, 4096]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 16384, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}}, "use_zarr3": false, "store_array_data_equal_to_fill_value": true, "custom_metadata": null} \ No newline at end of file diff --git a/5000/params/_sharding b/5000/params/_sharding new file mode 100644 index 0000000000000000000000000000000000000000..2df2a383e2412bac6c7c432773d4453584175688 --- /dev/null +++ b/5000/params/_sharding @@ -0,0 +1 @@ +{"cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcucG9zX2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZW1iZWRkZXIuaW5wdXRfZW1iZWRkaW5nLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZmluYWxfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5nYXRpbmdfZWluc3VtLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5saW5lYXIudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4uYXR0bl92ZWNfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ua3ZfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ucV9laW5zdW0udy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9hdHRlbnRpb25fbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9mZndfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}"} \ No newline at end of file diff --git a/5000/params/array_metadatas/process_0 b/5000/params/array_metadatas/process_0 new file mode 100644 index 0000000000000000000000000000000000000000..0c90735afe75d094cf4bcd644a082a83f3187232 --- /dev/null +++ b/5000/params/array_metadatas/process_0 @@ -0,0 +1 @@ +{"array_metadatas": [{"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 1076], "chunk_shape": [27, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 1152, 1076], "chunk_shape": [27, 1152, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 4304, 288], "chunk_shape": [27, 4304, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 16, 72, 288], "chunk_shape": [27, 16, 72, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 288], "chunk_shape": [14, 14, 3, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.bias.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.kernel.value", "write_shape": [1152, 512], "chunk_shape": [1152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.pos_embedding.value", "write_shape": [1, 64, 1152], "chunk_shape": [1, 64, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [257152, 512], "chunk_shape": [257152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.final_norm.scale.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 8, 256, 512], "chunk_shape": [18, 8, 256, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 2048, 64], "chunk_shape": [18, 2, 1, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 8, 2048, 64], "chunk_shape": [18, 8, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 2048, 4096], "chunk_shape": [18, 2, 2048, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 16384, 512], "chunk_shape": [18, 16384, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}]} \ No newline at end of file diff --git a/5000/params/d/c574986e5ea1aabc4f9ecee8b25ef7ea b/5000/params/d/c574986e5ea1aabc4f9ecee8b25ef7ea new file mode 100644 index 0000000000000000000000000000000000000000..4f337742d59aeb214427a5a02b1635fcb9cc97d5 Binary files /dev/null and b/5000/params/d/c574986e5ea1aabc4f9ecee8b25ef7ea differ diff --git a/5000/params/manifest.ocdbt b/5000/params/manifest.ocdbt new file mode 100644 index 0000000000000000000000000000000000000000..5b0d3af0215081c5c8761cf7b4abb435521b27a7 Binary files /dev/null and b/5000/params/manifest.ocdbt differ diff --git a/5000/params/ocdbt.process_0/d/25235ade121ee0cee4066e0d143ea666 b/5000/params/ocdbt.process_0/d/25235ade121ee0cee4066e0d143ea666 new file mode 100644 index 0000000000000000000000000000000000000000..181c5007bfd002712ce5e1471cc00a936efea5cc Binary files /dev/null and b/5000/params/ocdbt.process_0/d/25235ade121ee0cee4066e0d143ea666 differ diff --git a/5000/params/ocdbt.process_0/d/32df0fee70688bdd2ace647a64a18e8f b/5000/params/ocdbt.process_0/d/32df0fee70688bdd2ace647a64a18e8f new file mode 100644 index 0000000000000000000000000000000000000000..c9e8f40c482c7c565cbc8b45252e444c6bb8cfe7 --- /dev/null +++ b/5000/params/ocdbt.process_0/d/32df0fee70688bdd2ace647a64a18e8f @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3be5046754568e114213a632cd3ae7600cd6369faf9140de129e3294a929db44 +size 180103201 diff --git a/5000/params/ocdbt.process_0/d/5845fb67e1a7192373e6ca9f4bf6d4ce b/5000/params/ocdbt.process_0/d/5845fb67e1a7192373e6ca9f4bf6d4ce new file mode 100644 index 0000000000000000000000000000000000000000..95affbc557593370d43de97cdd0c08743f9cfae9 --- /dev/null +++ b/5000/params/ocdbt.process_0/d/5845fb67e1a7192373e6ca9f4bf6d4ce @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ccc16627e5ef5bdc52322c8c8792305ae39b1438bae72aa87d3ef94e42ccf4ed +size 575916845 diff --git a/5000/params/ocdbt.process_0/d/61480e7b5b517efc12786e7fa2ce2c9c b/5000/params/ocdbt.process_0/d/61480e7b5b517efc12786e7fa2ce2c9c new file mode 100644 index 0000000000000000000000000000000000000000..b046c97eb4663626c56adeab8a66807ad5faa426 --- /dev/null +++ b/5000/params/ocdbt.process_0/d/61480e7b5b517efc12786e7fa2ce2c9c @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5632cd9c82fd37da40c991613dcdad32142f7fdb47373be29900a1a118c8a3fd +size 2240653077 diff --git a/5000/params/ocdbt.process_0/d/63af9d7a36959845369d5a45aa17ad79 b/5000/params/ocdbt.process_0/d/63af9d7a36959845369d5a45aa17ad79 new file mode 100644 index 0000000000000000000000000000000000000000..9f7599a2202783e6afff94ad49024fbac8b6b2f7 --- /dev/null +++ b/5000/params/ocdbt.process_0/d/63af9d7a36959845369d5a45aa17ad79 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4b4661473641ce0711b54613fdb52593d6c896bb863e7aab399e94de88cb70d5 +size 267314 diff --git a/5000/params/ocdbt.process_0/d/aafa75fec409c41d9dca6ba33f0b6baa b/5000/params/ocdbt.process_0/d/aafa75fec409c41d9dca6ba33f0b6baa new file mode 100644 index 0000000000000000000000000000000000000000..b9791923e554f5917d9ffca79ce749ad5b504812 --- /dev/null +++ b/5000/params/ocdbt.process_0/d/aafa75fec409c41d9dca6ba33f0b6baa @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2fa943eb248248280ca51d0533988e1bd64fd0e6b361fb6425f8c84e1b493f7a +size 1414428852 diff --git a/5000/params/ocdbt.process_0/d/b11c9e7adca87a7f80ef850730d5396d b/5000/params/ocdbt.process_0/d/b11c9e7adca87a7f80ef850730d5396d new file mode 100644 index 0000000000000000000000000000000000000000..159dc934828f74c7e73d7a7ee23ef21027351374 Binary files /dev/null and b/5000/params/ocdbt.process_0/d/b11c9e7adca87a7f80ef850730d5396d differ diff --git a/5000/params/ocdbt.process_0/d/c7e602a5c47b7bf6004c22d44ab961e8 b/5000/params/ocdbt.process_0/d/c7e602a5c47b7bf6004c22d44ab961e8 new file mode 100644 index 0000000000000000000000000000000000000000..b1032967f80ba42d38383bd33f354f993a957e70 Binary files /dev/null and b/5000/params/ocdbt.process_0/d/c7e602a5c47b7bf6004c22d44ab961e8 differ diff --git a/5000/params/ocdbt.process_0/d/c9e38940c1863fa768c2684ff7c0c09e b/5000/params/ocdbt.process_0/d/c9e38940c1863fa768c2684ff7c0c09e new file mode 100644 index 0000000000000000000000000000000000000000..504310fc9cb7dc7e80471d8bfa9e80458868f08b --- /dev/null +++ b/5000/params/ocdbt.process_0/d/c9e38940c1863fa768c2684ff7c0c09e @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:18b96c8dff907c983d827dc1fc1333e6565810fc335a92ddeba498c675839ade +size 1680766724 diff --git a/5000/params/ocdbt.process_0/d/d6b85ffe71d6c1ae0f70155f5248eb05 b/5000/params/ocdbt.process_0/d/d6b85ffe71d6c1ae0f70155f5248eb05 new file mode 100644 index 0000000000000000000000000000000000000000..f7bbcdd9204a8dc89cc16ad0a17c56871d82f91e --- /dev/null +++ b/5000/params/ocdbt.process_0/d/d6b85ffe71d6c1ae0f70155f5248eb05 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f431c82279f2c8b16658a90905be28cf53c7a2082d858d67959c7a4d923bc815 +size 2517756210 diff --git a/5000/params/ocdbt.process_0/d/fc11f1f051eac1bbf8cd0c6a03a62a41 b/5000/params/ocdbt.process_0/d/fc11f1f051eac1bbf8cd0c6a03a62a41 new file mode 100644 index 0000000000000000000000000000000000000000..2dfb750ec8cf9bb4ed0006082ca7b570d6cde5e6 --- /dev/null +++ b/5000/params/ocdbt.process_0/d/fc11f1f051eac1bbf8cd0c6a03a62a41 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:73fcefda70dc18d66f341a4f220c8da16203a24d1db21397a34207a067a962ec +size 2240650719 diff --git a/5000/params/ocdbt.process_0/manifest.ocdbt b/5000/params/ocdbt.process_0/manifest.ocdbt new file mode 100644 index 0000000000000000000000000000000000000000..6245af89be4be5bfe95fd1e155ce5b8df0705634 Binary files /dev/null and b/5000/params/ocdbt.process_0/manifest.ocdbt differ diff --git a/5000/train_state/_METADATA b/5000/train_state/_METADATA new file mode 100644 index 0000000000000000000000000000000000000000..ec9c7af7a310de1fdbd88a0fb82a39b7f5dd4af9 --- /dev/null +++ b/5000/train_state/_METADATA @@ -0,0 +1 @@ +{"tree_metadata": {"('step',)": {"key_metadata": [{"key": "step", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": []}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1076]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 1076]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4304, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 16, 72, 288]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('params', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('params', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 288]}}, "('params', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('params', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1152, 512]}}, "('params', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 64, 1152]}}, "('params', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [257152, 512]}}, "('params', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 256, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2048, 64]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 2048, 64]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048, 4096]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 16384, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('model_def',)": {"key_metadata": [{"key": "model_def", "key_type": 2}], "value_metadata": {"value_type": "None", "skip_deserialize": true}}, "('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', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1076]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 1076]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4304, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 16, 72, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 288]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1152, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 64, 1152]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [257152, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 256, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2048, 64]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 2048, 64]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048, 4096]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 16384, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('opt_state', '1', '0', 'mu', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "mu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1076]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 1076]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4304, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 16, 72, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 4, 72]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 1152, 16, 18]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 288]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1152, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 64, 1152]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [257152, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 256, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2048, 64]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 8, 2048, 64]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048, 4096]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 16384, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('opt_state', '1', '0', 'nu', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "0", "key_type": 1}, {"key": "nu", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512]}}, "('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', 'count')": {"key_metadata": [{"key": "opt_state", "key_type": 2}, {"key": "1", "key_type": 1}, {"key": "2", "key_type": 1}, {"key": "count", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": []}}, "('ema_params',)": {"key_metadata": [{"key": "ema_params", "key_type": 2}], "value_metadata": {"value_type": "None", "skip_deserialize": true}}}, "use_zarr3": false, "store_array_data_equal_to_fill_value": true, "custom_metadata": null} \ No newline at end of file diff --git a/5000/train_state/_sharding b/5000/train_state/_sharding new file mode 100644 index 0000000000000000000000000000000000000000..a7a7ba4fbfbe4420b2930de10b935d2c467ee425 --- /dev/null +++ b/5000/train_state/_sharding @@ -0,0 +1 @@ +{"b3B0X3N0YXRlLjEuMC5jb3VudA==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJfbm9ybS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMC5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzAuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzAua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzEuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzEua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAua2V5LmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAua2V5Lmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAub3V0LmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAub3V0Lmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAucXVlcnkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAucXVlcnkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAudmFsdWUuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAudmFsdWUua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLmVtYmVkZGluZy5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLmVtYmVkZGluZy5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLmhlYWQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLmhlYWQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEuaW1nLnBvc19lbWJlZGRpbmcudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmVtYmVkZGVyLmlucHV0X2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmZpbmFsX25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLmF0dG5fdmVjX2VpbnN1bS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLmt2X2VpbnN1bS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLnFfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5tbHAuZ2F0aW5nX2VpbnN1bS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5tbHAubGluZWFyLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5wcmVfYXR0ZW50aW9uX25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5tdS5QYWxpR2VtbWEubGxtLmxheWVycy5wcmVfZmZ3X25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJfbm9ybS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMC5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5MYXllck5vcm1fMS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzAuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzAua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzEuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NbHBCbG9ja18wLkRlbnNlXzEua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAua2V5LmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAua2V5Lmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAub3V0LmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAub3V0Lmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAucXVlcnkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAucXVlcnkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAudmFsdWUuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLlRyYW5zZm9ybWVyLmVuY29kZXJibG9jay5NdWx0aUhlYWREb3RQcm9kdWN0QXR0ZW50aW9uXzAudmFsdWUua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLmVtYmVkZGluZy5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLmVtYmVkZGluZy5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLmhlYWQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLmhlYWQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEuaW1nLnBvc19lbWJlZGRpbmcudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmVtYmVkZGVyLmlucHV0X2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmZpbmFsX25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLmF0dG5fdmVjX2VpbnN1bS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLmt2X2VpbnN1bS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5hdHRuLnFfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5tbHAuZ2F0aW5nX2VpbnN1bS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5tbHAubGluZWFyLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5wcmVfYXR0ZW50aW9uX25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMC5udS5QYWxpR2VtbWEubGxtLmxheWVycy5wcmVfZmZ3X25vcm0uc2NhbGUudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","b3B0X3N0YXRlLjEuMi5jb3VudA==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","c3RlcA==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcucG9zX2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZW1iZWRkZXIuaW5wdXRfZW1iZWRkaW5nLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZmluYWxfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5nYXRpbmdfZWluc3VtLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5saW5lYXIudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4uYXR0bl92ZWNfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ua3ZfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ucV9laW5zdW0udy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [null, null, null, \"fsdp\"], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9hdHRlbnRpb25fbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9mZndfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [1, 4], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}, {\"id\": 1}, {\"id\": 2}, {\"id\": 3}]]}}"} \ No newline at end of file diff --git a/5000/train_state/array_metadatas/process_0 b/5000/train_state/array_metadatas/process_0 new file mode 100644 index 0000000000000000000000000000000000000000..9a1f979d0388a8b0b34ba00b6e06c02a0127ff69 --- /dev/null +++ b/5000/train_state/array_metadatas/process_0 @@ -0,0 +1 @@ +{"array_metadatas": [{"array_metadata": {"param_name": "step", "write_shape": [], "chunk_shape": [], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 1076], "chunk_shape": [27, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 1152, 1076], "chunk_shape": [27, 1152, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 4304, 288], "chunk_shape": [27, 4304, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 16, 72, 288], "chunk_shape": [27, 16, 72, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 288], "chunk_shape": [14, 14, 3, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.bias.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.kernel.value", "write_shape": [1152, 512], "chunk_shape": [1152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.pos_embedding.value", "write_shape": [1, 64, 1152], "chunk_shape": [1, 64, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [257152, 512], "chunk_shape": [257152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.final_norm.scale.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 8, 256, 512], "chunk_shape": [18, 8, 256, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 2048, 64], "chunk_shape": [18, 2, 1, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 8, 2048, 64], "chunk_shape": [18, 8, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 2048, 4096], "chunk_shape": [18, 2, 2048, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 16384, 512], "chunk_shape": [18, 16384, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "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.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 1076], "chunk_shape": [27, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 1152, 1076], "chunk_shape": [27, 1152, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 4304, 288], "chunk_shape": [27, 4304, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 16, 72, 288], "chunk_shape": [27, 16, 72, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.embedding.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 288], "chunk_shape": [14, 14, 3, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.head.bias.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.head.kernel.value", "write_shape": [1152, 512], "chunk_shape": [1152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.pos_embedding.value", "write_shape": [1, 64, 1152], "chunk_shape": [1, 64, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [257152, 512], "chunk_shape": [257152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.final_norm.scale.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 8, 256, 512], "chunk_shape": [18, 8, 256, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 2048, 64], "chunk_shape": [18, 2, 1, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 8, 2048, 64], "chunk_shape": [18, 8, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 2048, 4096], "chunk_shape": [18, 2, 2048, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 16384, 512], "chunk_shape": [18, 16384, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 1076], "chunk_shape": [27, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 1152, 1076], "chunk_shape": [27, 1152, 1076], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 4304, 288], "chunk_shape": [27, 4304, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 288], "chunk_shape": [27, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 16, 72, 288], "chunk_shape": [27, 16, 72, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 4, 72], "chunk_shape": [27, 4, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 1152, 16, 18], "chunk_shape": [27, 1152, 16, 18], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.embedding.bias.value", "write_shape": [288], "chunk_shape": [288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 288], "chunk_shape": [14, 14, 3, 288], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.head.bias.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.head.kernel.value", "write_shape": [1152, 512], "chunk_shape": [1152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.pos_embedding.value", "write_shape": [1, 64, 1152], "chunk_shape": [1, 64, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [257152, 512], "chunk_shape": [257152, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.final_norm.scale.value", "write_shape": [512], "chunk_shape": [512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 8, 256, 512], "chunk_shape": [18, 8, 256, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 2048, 64], "chunk_shape": [18, 2, 1, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 8, 2048, 64], "chunk_shape": [18, 8, 2048, 64], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 2048, 4096], "chunk_shape": [18, 2, 2048, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 16384, 512], "chunk_shape": [18, 16384, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 512], "chunk_shape": [18, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.2.count", "write_shape": [], "chunk_shape": [], "ext_metadata": null}}]} \ No newline at end of file diff --git a/5000/train_state/d/d36dcb88abdb96db80d7fe4c7952e707 b/5000/train_state/d/d36dcb88abdb96db80d7fe4c7952e707 new file mode 100644 index 0000000000000000000000000000000000000000..9480fb0fde577926c3295887480053a9ee121edb Binary files /dev/null and b/5000/train_state/d/d36dcb88abdb96db80d7fe4c7952e707 differ diff --git a/5000/train_state/manifest.ocdbt b/5000/train_state/manifest.ocdbt new file mode 100644 index 0000000000000000000000000000000000000000..dc9bea15b3f1375665300f55c36d95f5f2127eba Binary files /dev/null and b/5000/train_state/manifest.ocdbt differ diff --git a/5000/train_state/ocdbt.process_0/d/0f5181f9ff451b8a8f8ff9679c8cb5c6 b/5000/train_state/ocdbt.process_0/d/0f5181f9ff451b8a8f8ff9679c8cb5c6 new file mode 100644 index 0000000000000000000000000000000000000000..5841b89883314c036adae2b0539c307828397602 --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/0f5181f9ff451b8a8f8ff9679c8cb5c6 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:374eed7b8ac9950ca9da112b04954a0752290f35e92767da5371333e95571cf6 +size 2671816084 diff --git a/5000/train_state/ocdbt.process_0/d/39298a92eda90c6cbcd86e5c6be1bd33 b/5000/train_state/ocdbt.process_0/d/39298a92eda90c6cbcd86e5c6be1bd33 new file mode 100644 index 0000000000000000000000000000000000000000..a3c6636a0b3da525a9205b802767c1fe221e60f4 --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/39298a92eda90c6cbcd86e5c6be1bd33 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cee0d050b0417815709cd69fbe0c732444680274212c2de86dc944f8567d1d3b +size 2155547137 diff --git a/5000/train_state/ocdbt.process_0/d/3f132fa253cbe9cdb3fedb6523babd90 b/5000/train_state/ocdbt.process_0/d/3f132fa253cbe9cdb3fedb6523babd90 new file mode 100644 index 0000000000000000000000000000000000000000..aa5047a4ecfcf130a5f79f80a075a2975f1c98f9 --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/3f132fa253cbe9cdb3fedb6523babd90 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a63b3762952210c829fcd31355d7ac17a877f18386ae58559b28049f56727228 +size 2176248015 diff --git a/5000/train_state/ocdbt.process_0/d/415d9126b07405df52be070cb0ac638a b/5000/train_state/ocdbt.process_0/d/415d9126b07405df52be070cb0ac638a new file mode 100644 index 0000000000000000000000000000000000000000..e527cbdec6d6c1f2b8fe0062afd7f462400ae466 --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/415d9126b07405df52be070cb0ac638a @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f308b65e9974c6d5cb63bcd0378fee71761de3fcc6671ba500b24ead0762cf0 +size 2179721253 diff --git a/5000/train_state/ocdbt.process_0/d/5f336b2a983cdcad11aa83cbecc89b61 b/5000/train_state/ocdbt.process_0/d/5f336b2a983cdcad11aa83cbecc89b61 new file mode 100644 index 0000000000000000000000000000000000000000..65543f15db2e57c4d1d633780e0cf173336a164c --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/5f336b2a983cdcad11aa83cbecc89b61 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:af3c1b67d723f1dd41a02069e303e14160214312339d8cdf8d49d1690e944561 +size 2229763099 diff --git a/5000/train_state/ocdbt.process_0/d/6524a676f960e14b02cda3de5b0ef389 b/5000/train_state/ocdbt.process_0/d/6524a676f960e14b02cda3de5b0ef389 new file mode 100644 index 0000000000000000000000000000000000000000..a9e69151ed6f275a26c0f30c8d41212162963267 Binary files /dev/null and b/5000/train_state/ocdbt.process_0/d/6524a676f960e14b02cda3de5b0ef389 differ diff --git a/5000/train_state/ocdbt.process_0/d/6dc76498ed5d71fbb6f96e36291d0f2f b/5000/train_state/ocdbt.process_0/d/6dc76498ed5d71fbb6f96e36291d0f2f new file mode 100644 index 0000000000000000000000000000000000000000..15aaae2455bbb76aad4af551b7ecedd4c25aaf6b Binary files /dev/null and b/5000/train_state/ocdbt.process_0/d/6dc76498ed5d71fbb6f96e36291d0f2f differ diff --git a/5000/train_state/ocdbt.process_0/d/72e228cb96f9e0070f73b720eda6f9b1 b/5000/train_state/ocdbt.process_0/d/72e228cb96f9e0070f73b720eda6f9b1 new file mode 100644 index 0000000000000000000000000000000000000000..5d89f5d25c262ac36af570464601c3d0e77352c2 --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/72e228cb96f9e0070f73b720eda6f9b1 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d91848b78b3f6e0a655bf0d79616f4aa77bfb8a27ad1afa7b94d1746d9625b4f +size 2465616045 diff --git a/5000/train_state/ocdbt.process_0/d/7867b5092afd15dd321c5852192561c2 b/5000/train_state/ocdbt.process_0/d/7867b5092afd15dd321c5852192561c2 new file mode 100644 index 0000000000000000000000000000000000000000..aa1f93de1e3801064ee3f4bf58bfcf21bf9d3949 --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/7867b5092afd15dd321c5852192561c2 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6efa3c246c1c2dcdcdaf1f86108530b3afd60e10bd90473a78373402275c4d55 +size 2246775760 diff --git a/5000/train_state/ocdbt.process_0/d/806e8a76755eacea75950df8d2b87476 b/5000/train_state/ocdbt.process_0/d/806e8a76755eacea75950df8d2b87476 new file mode 100644 index 0000000000000000000000000000000000000000..ac020a1dc6a63fbdc7761621a6f0902228427bed Binary files /dev/null and b/5000/train_state/ocdbt.process_0/d/806e8a76755eacea75950df8d2b87476 differ diff --git a/5000/train_state/ocdbt.process_0/d/834d2ee970214790e27b5a5cc2ea94cd b/5000/train_state/ocdbt.process_0/d/834d2ee970214790e27b5a5cc2ea94cd new file mode 100644 index 0000000000000000000000000000000000000000..eb9848367dba0128f931b7ee00008cdbf7c70513 --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/834d2ee970214790e27b5a5cc2ea94cd @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d8863874de8d9f9a3506f35fdaa897cc2ad4bec9f2e2b81dce969fed2ddc261 +size 2167609490 diff --git a/5000/train_state/ocdbt.process_0/d/84e79316947a97b7fff3f2764b985102 b/5000/train_state/ocdbt.process_0/d/84e79316947a97b7fff3f2764b985102 new file mode 100644 index 0000000000000000000000000000000000000000..44bdce85658d97e157a4c96509ccdf50068e7d52 --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/84e79316947a97b7fff3f2764b985102 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aad07265b82848cb37188be70f042c2c9b52704802c94257bde2f85c8f04104b +size 2211534897 diff --git a/5000/train_state/ocdbt.process_0/d/84f9865b68b638ced9fac9a5a883fed2 b/5000/train_state/ocdbt.process_0/d/84f9865b68b638ced9fac9a5a883fed2 new file mode 100644 index 0000000000000000000000000000000000000000..cffce5ffa3afdfd3c7ec6dd393f8b3d8e0148574 --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/84f9865b68b638ced9fac9a5a883fed2 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b95b96867b3761996b4f5bd4508e829ad6ff292e278b0b0e543a4114daf8810e +size 116753 diff --git a/5000/train_state/ocdbt.process_0/d/930b7ac293689f98bf1a50f236294336 b/5000/train_state/ocdbt.process_0/d/930b7ac293689f98bf1a50f236294336 new file mode 100644 index 0000000000000000000000000000000000000000..eb838aa7cf21dd4465365f9487b212ad93a831f1 Binary files /dev/null and b/5000/train_state/ocdbt.process_0/d/930b7ac293689f98bf1a50f236294336 differ diff --git a/5000/train_state/ocdbt.process_0/d/9b14aaab37cfecc4d31e1e521c555f86 b/5000/train_state/ocdbt.process_0/d/9b14aaab37cfecc4d31e1e521c555f86 new file mode 100644 index 0000000000000000000000000000000000000000..7b1ac9469b646bc4ce186ab2bdd53584facad9df --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/9b14aaab37cfecc4d31e1e521c555f86 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2416ad8116ebd869d2be128c8b5d8bf8d875a26350c62bafb200a78ee64898fe +size 2157961616 diff --git a/5000/train_state/ocdbt.process_0/d/a7cae5a53d124d28c8138c9fd72c47c7 b/5000/train_state/ocdbt.process_0/d/a7cae5a53d124d28c8138c9fd72c47c7 new file mode 100644 index 0000000000000000000000000000000000000000..0680b41b6cbd626c6d05a44dd14a86a7a859c5aa --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/a7cae5a53d124d28c8138c9fd72c47c7 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c80197d6fcc85ca3db2b2425b2a07e4f3139229c313b3c90e40620f6b5f91645 +size 2210317429 diff --git a/5000/train_state/ocdbt.process_0/d/c4e019d978d4f8260f94ab8f9ac38914 b/5000/train_state/ocdbt.process_0/d/c4e019d978d4f8260f94ab8f9ac38914 new file mode 100644 index 0000000000000000000000000000000000000000..723c4196cdfd9c52981f9faf9cc90377d3a3155c --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/c4e019d978d4f8260f94ab8f9ac38914 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:21d57218fd3852dc2f072bcb496d3b2d122c0c65c975f353ab3df63f20a15958 +size 2758611585 diff --git a/5000/train_state/ocdbt.process_0/d/d53ca3e9a6142c2135930446284c0a4c b/5000/train_state/ocdbt.process_0/d/d53ca3e9a6142c2135930446284c0a4c new file mode 100644 index 0000000000000000000000000000000000000000..3bb5f328f7cd4428df99ad27fde0d5d49db107a4 --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/d53ca3e9a6142c2135930446284c0a4c @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4c71cb5584b4a337571a297bc4a75add5bf206c13c91ddfbf98c4b776984bf80 +size 2240650108 diff --git a/5000/train_state/ocdbt.process_0/d/e454b21615471a5f1ecb3c59b80ecd4a b/5000/train_state/ocdbt.process_0/d/e454b21615471a5f1ecb3c59b80ecd4a new file mode 100644 index 0000000000000000000000000000000000000000..5d148e64f2eee73db5fcefde353990a5cce7558a --- /dev/null +++ b/5000/train_state/ocdbt.process_0/d/e454b21615471a5f1ecb3c59b80ecd4a @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8397885692d5fe9011ea3b912fc3759a5730e9115418bfd4f0dd5d873e547812 +size 2243790277 diff --git a/5000/train_state/ocdbt.process_0/d/f94de6c7fd7836379a673fa7eb4cfb19 b/5000/train_state/ocdbt.process_0/d/f94de6c7fd7836379a673fa7eb4cfb19 new file mode 100644 index 0000000000000000000000000000000000000000..6c951cf616c88514eece99d1b9117c1a37682361 Binary files /dev/null and b/5000/train_state/ocdbt.process_0/d/f94de6c7fd7836379a673fa7eb4cfb19 differ diff --git a/5000/train_state/ocdbt.process_0/manifest.ocdbt b/5000/train_state/ocdbt.process_0/manifest.ocdbt new file mode 100644 index 0000000000000000000000000000000000000000..e9ff32a4a97a75b8358eb1134fac5083d9c41c8f Binary files /dev/null and b/5000/train_state/ocdbt.process_0/manifest.ocdbt differ diff --git a/wandb_id.txt b/wandb_id.txt new file mode 100644 index 0000000000000000000000000000000000000000..d2e39fa73b0b29335822d3f62102977debe0d344 --- /dev/null +++ b/wandb_id.txt @@ -0,0 +1 @@ +ehocn154 \ No newline at end of file