mt-prox's picture
epoch 1/5: val_loss=5.1989 val_acc=0.008
7e00590 verified
Raw
History Blame Contribute Delete
20.1 kB
{"array_metadatas": [{"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoder_norm.bias", "write_shape": [1152], "chunk_shape": [1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoder_norm.scale", "write_shape": [1152], "chunk_shape": [1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias", "write_shape": [27, 4304], "chunk_shape": [27, 4304], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel", "write_shape": [27, 1152, 4304], "chunk_shape": [27, 1152, 4304], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel", "write_shape": [27, 4304, 1152], "chunk_shape": [27, 4304, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias", "write_shape": [27, 16, 72], "chunk_shape": [27, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel", "write_shape": [27, 1152, 16, 72], "chunk_shape": [27, 1152, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel", "write_shape": [27, 16, 72, 1152], "chunk_shape": [27, 16, 72, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias", "write_shape": [27, 16, 72], "chunk_shape": [27, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel", "write_shape": [27, 1152, 16, 72], "chunk_shape": [27, 1152, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias", "write_shape": [27, 16, 72], "chunk_shape": [27, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel", "write_shape": [27, 1152, 16, 72], "chunk_shape": [27, 1152, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.embedding.bias", "write_shape": [1152], "chunk_shape": [1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.embedding.kernel", "write_shape": [14, 14, 3, 1152], "chunk_shape": [14, 14, 3, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.head.bias", "write_shape": [1024], "chunk_shape": [1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.head.kernel", "write_shape": [1152, 1024], "chunk_shape": [1152, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.img.pos_embedding", "write_shape": [1, 256, 1152], "chunk_shape": [1, 256, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.llm.embedder.input_embedding", "write_shape": [257152, 1024], "chunk_shape": [257152, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.llm.final_norm.scale", "write_shape": [1024], "chunk_shape": [1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.llm.layers.attn.attn_vec_einsum.w", "write_shape": [18, 8, 256, 1024], "chunk_shape": [18, 8, 256, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.llm.layers.attn.kv_einsum.w", "write_shape": [18, 2, 1, 1024, 256], "chunk_shape": [18, 2, 1, 1024, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.llm.layers.attn.q_einsum.w", "write_shape": [18, 8, 1024, 256], "chunk_shape": [18, 8, 1024, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.llm.layers.mlp.gating_einsum", "write_shape": [18, 2, 1024, 4096], "chunk_shape": [18, 2, 1024, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.llm.layers.mlp.linear", "write_shape": [18, 4096, 1024], "chunk_shape": [18, 4096, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.llm.layers.pre_attention_norm.scale", "write_shape": [18, 1024], "chunk_shape": [18, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.PaliGemma.llm.layers.pre_ffw_norm.scale", "write_shape": [18, 1024], "chunk_shape": [18, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "model._value_head_linears.0.bias", "write_shape": [1024], "chunk_shape": [1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "model._value_head_linears.0.kernel", "write_shape": [1024, 1024], "chunk_shape": [1024, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.value_head_out.bias", "write_shape": [201], "chunk_shape": [201], "ext_metadata": null}}, {"array_metadata": {"param_name": "model.value_head_out.kernel", "write_shape": [1024, 201], "chunk_shape": [1024, 201], "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", "write_shape": [1152], "chunk_shape": [1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoder_norm.scale", "write_shape": [1152], "chunk_shape": [1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias", "write_shape": [27, 4304], "chunk_shape": [27, 4304], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel", "write_shape": [27, 1152, 4304], "chunk_shape": [27, 1152, 4304], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel", "write_shape": [27, 4304, 1152], "chunk_shape": [27, 4304, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias", "write_shape": [27, 16, 72], "chunk_shape": [27, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel", "write_shape": [27, 1152, 16, 72], "chunk_shape": [27, 1152, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel", "write_shape": [27, 16, 72, 1152], "chunk_shape": [27, 16, 72, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias", "write_shape": [27, 16, 72], "chunk_shape": [27, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel", "write_shape": [27, 1152, 16, 72], "chunk_shape": [27, 1152, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias", "write_shape": [27, 16, 72], "chunk_shape": [27, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel", "write_shape": [27, 1152, 16, 72], "chunk_shape": [27, 1152, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.embedding.bias", "write_shape": [1152], "chunk_shape": [1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.embedding.kernel", "write_shape": [14, 14, 3, 1152], "chunk_shape": [14, 14, 3, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.head.bias", "write_shape": [1024], "chunk_shape": [1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.head.kernel", "write_shape": [1152, 1024], "chunk_shape": [1152, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.img.pos_embedding", "write_shape": [1, 256, 1152], "chunk_shape": [1, 256, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.embedder.input_embedding", "write_shape": [257152, 1024], "chunk_shape": [257152, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.final_norm.scale", "write_shape": [1024], "chunk_shape": [1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.attn.attn_vec_einsum.w", "write_shape": [18, 8, 256, 1024], "chunk_shape": [18, 8, 256, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.attn.kv_einsum.w", "write_shape": [18, 2, 1, 1024, 256], "chunk_shape": [18, 2, 1, 1024, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.attn.q_einsum.w", "write_shape": [18, 8, 1024, 256], "chunk_shape": [18, 8, 1024, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.mlp.gating_einsum", "write_shape": [18, 2, 1024, 4096], "chunk_shape": [18, 2, 1024, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.mlp.linear", "write_shape": [18, 4096, 1024], "chunk_shape": [18, 4096, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.pre_attention_norm.scale", "write_shape": [18, 1024], "chunk_shape": [18, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.PaliGemma.llm.layers.pre_ffw_norm.scale", "write_shape": [18, 1024], "chunk_shape": [18, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu._value_head_linears.0.bias", "write_shape": [1024], "chunk_shape": [1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu._value_head_linears.0.kernel", "write_shape": [1024, 1024], "chunk_shape": [1024, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.value_head_out.bias", "write_shape": [201], "chunk_shape": [201], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.mu.value_head_out.kernel", "write_shape": [1024, 201], "chunk_shape": [1024, 201], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoder_norm.bias", "write_shape": [1152], "chunk_shape": [1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoder_norm.scale", "write_shape": [1152], "chunk_shape": [1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias", "write_shape": [27, 4304], "chunk_shape": [27, 4304], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel", "write_shape": [27, 1152, 4304], "chunk_shape": [27, 1152, 4304], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel", "write_shape": [27, 4304, 1152], "chunk_shape": [27, 4304, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias", "write_shape": [27, 16, 72], "chunk_shape": [27, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel", "write_shape": [27, 1152, 16, 72], "chunk_shape": [27, 1152, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias", "write_shape": [27, 1152], "chunk_shape": [27, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel", "write_shape": [27, 16, 72, 1152], "chunk_shape": [27, 16, 72, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias", "write_shape": [27, 16, 72], "chunk_shape": [27, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel", "write_shape": [27, 1152, 16, 72], "chunk_shape": [27, 1152, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias", "write_shape": [27, 16, 72], "chunk_shape": [27, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel", "write_shape": [27, 1152, 16, 72], "chunk_shape": [27, 1152, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.embedding.bias", "write_shape": [1152], "chunk_shape": [1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.embedding.kernel", "write_shape": [14, 14, 3, 1152], "chunk_shape": [14, 14, 3, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.head.bias", "write_shape": [1024], "chunk_shape": [1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.head.kernel", "write_shape": [1152, 1024], "chunk_shape": [1152, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.img.pos_embedding", "write_shape": [1, 256, 1152], "chunk_shape": [1, 256, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.embedder.input_embedding", "write_shape": [257152, 1024], "chunk_shape": [257152, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.final_norm.scale", "write_shape": [1024], "chunk_shape": [1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.attn.attn_vec_einsum.w", "write_shape": [18, 8, 256, 1024], "chunk_shape": [18, 8, 256, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.attn.kv_einsum.w", "write_shape": [18, 2, 1, 1024, 256], "chunk_shape": [18, 2, 1, 1024, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.attn.q_einsum.w", "write_shape": [18, 8, 1024, 256], "chunk_shape": [18, 8, 1024, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.mlp.gating_einsum", "write_shape": [18, 2, 1024, 4096], "chunk_shape": [18, 2, 1024, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.mlp.linear", "write_shape": [18, 4096, 1024], "chunk_shape": [18, 4096, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.pre_attention_norm.scale", "write_shape": [18, 1024], "chunk_shape": [18, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.PaliGemma.llm.layers.pre_ffw_norm.scale", "write_shape": [18, 1024], "chunk_shape": [18, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu._value_head_linears.0.bias", "write_shape": [1024], "chunk_shape": [1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu._value_head_linears.0.kernel", "write_shape": [1024, 1024], "chunk_shape": [1024, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.value_head_out.bias", "write_shape": [201], "chunk_shape": [201], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.0.nu.value_head_out.kernel", "write_shape": [1024, 201], "chunk_shape": [1024, 201], "ext_metadata": null}}, {"array_metadata": {"param_name": "opt_state.1.2.count", "write_shape": [], "chunk_shape": [], "ext_metadata": null}}, {"array_metadata": {"param_name": "step", "write_shape": [], "chunk_shape": [], "ext_metadata": null}}]}