- metrics._reduction_counter.value
- metrics.loss.count
- metrics.loss.total
- opt_state.1.0.count
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.LayerNorm_0.bias
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.LayerNorm_0.scale
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.LayerNorm_1.bias
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.LayerNorm_1.scale
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.MLPBlock_0.Dense_0.bias
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.MLPBlock_0.Dense_0.kernel
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.MLPBlock_0.Dense_1.bias
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.MLPBlock_0.Dense_1.kernel
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.MultiHeadDotProductAttention_0.key.bias
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.MultiHeadDotProductAttention_0.key.kernel
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.MultiHeadDotProductAttention_0.out.bias
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.MultiHeadDotProductAttention_0.out.kernel
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.MultiHeadDotProductAttention_0.query.bias
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.MultiHeadDotProductAttention_0.query.kernel
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.MultiHeadDotProductAttention_0.value.bias
- opt_state.1.0.mu.attention_blocks.ScanEncoder1DBlock_0.MultiHeadDotProductAttention_0.value.kernel
- opt_state.1.0.mu.attention_blocks.posembed_input.pos_embedding
- opt_state.1.0.mu.categorical_action_head.Dense_0.bias
- opt_state.1.0.mu.categorical_action_head.Dense_0.kernel
- opt_state.1.0.mu.image_encoder.embedding_function.Conv_0.bias
- opt_state.1.0.mu.image_encoder.embedding_function.Conv_0.kernel
-
44.6 kB