diff --git a/.gitattributes b/.gitattributes index 4e898678400a8a45791d03b9a4e0fb8eba98aba8..eba33d1e40f1816b9b8d2a404d480626046e1b90 100644 --- a/.gitattributes +++ b/.gitattributes @@ -561,3 +561,253 @@ checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layer checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_4.mlp.wi_0.kernel/0.1 filter=lfs diff=lfs merge=lfs -text checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_5.encoder_decoder_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_19.encoder_decoder_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_8.attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_8.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_19.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_8.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_5.encoder_decoder_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_19.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_9.encoder_decoder_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_9.encoder_decoder_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_15.encoder_decoder_attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_0.encoder_decoder_attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_0.encoder_decoder_attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_8.mlp.wi_0.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_22.attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_8.mlp.wi_0.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_6.self_attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_22.attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_15.encoder_decoder_attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_8.mlp.wi_0.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_8.mlp.wi_0.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_6.self_attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_8.attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_20.attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_20.attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_0.self_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_8.attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_0.self_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_0.encoder_decoder_attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_0.encoder_decoder_attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_1.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_1.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_0.attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_0.attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_0.attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_4.encoder_decoder_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_0.attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_21.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_13.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_21.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_4.encoder_decoder_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_13.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_2.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_2.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_11.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_11.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_14.encoder_decoder_attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_14.encoder_decoder_attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_8.self_attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_8.self_attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_18.encoder_decoder_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_14.attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_18.encoder_decoder_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_0.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_0.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_19.mlp.wi_0.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_19.mlp.wi_0.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_7.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_21.attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_22.attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_21.attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_23.attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_22.attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_9.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_9.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_23.attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_9.encoder_decoder_attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_9.encoder_decoder_attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_23.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_3.encoder_decoder_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_23.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_3.encoder_decoder_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_7.self_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_7.self_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_12.attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_12.attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_4.attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_4.attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_20.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_20.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_18.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_18.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_12.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_23.mlp.wi_0.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_12.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_23.mlp.wi_0.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_0.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_0.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_1.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_1.self_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_1.self_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_19.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_1.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_19.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_11.encoder_decoder_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_18.self_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_3.mlp.wi_0.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_3.mlp.wi_0.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_18.self_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_2.encoder_decoder_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_13.attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_13.attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_2.encoder_decoder_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_10.mlp.wi_0.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_15.encoder_decoder_attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_11.self_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_11.self_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_10.mlp.wi_0.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_4.encoder_decoder_attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_0.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_4.encoder_decoder_attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_0.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_15.encoder_decoder_attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_8.encoder_decoder_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_8.encoder_decoder_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_15.self_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_11.encoder_decoder_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_15.self_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_4.attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_12.encoder_decoder_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_4.attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_12.encoder_decoder_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_3.attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_21.attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_21.attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_0.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_2.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_2.mlp.wi_0.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_18.attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_18.attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_2.mlp.wi_0.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_3.attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_0.self_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_0.self_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_19.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_19.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_0.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_7.encoder_decoder_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_7.encoder_decoder_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_14.encoder_decoder_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_5.encoder_decoder_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_3.attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_3.attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_5.encoder_decoder_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_11.self_attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_18.encoder_decoder_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_18.encoder_decoder_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_10.self_attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_6.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_11.self_attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_6.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_10.self_attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_5.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_5.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_2.self_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_2.self_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_17.encoder_decoder_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_17.encoder_decoder_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_3.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_3.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_19.self_attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_19.self_attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_14.self_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_4.mlp.wi_0.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_4.mlp.wi_0.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_15.attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_15.attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_13.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_13.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_18.attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_11.attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_11.attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_6.encoder_decoder_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_6.encoder_decoder_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_6.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_6.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_18.attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_13.encoder_decoder_attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_14.encoder_decoder_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_13.encoder_decoder_attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_9.encoder_decoder_attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_9.encoder_decoder_attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_14.self_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_5.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_5.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_3.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_3.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_15.attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_15.attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_11.self_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_19.self_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_0.self_attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_0.self_attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_8.self_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_0.self_attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_8.self_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_11.self_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_15.encoder_decoder_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_0.self_attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_15.encoder_decoder_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_16.attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_8.encoder_decoder_attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_8.encoder_decoder_attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_1.self_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_16.attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_19.self_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_1.self_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_11.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_11.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.logits_dense.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.logits_dense.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_20.attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_7.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_20.attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_10.mlp.wi_1.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_3.self_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_3.self_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_5.attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_5.attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_0.self_attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_5.attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_0.self_attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_10.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_11.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_11.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_7.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_8.mlp.wi_1.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_3.self_attention.out.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_1.mlp.wo.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_1.mlp.wo.kernel/1.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_3.self_attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_21.mlp.wi_0.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_18.attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_18.attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_21.mlp.wi_0.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_2.self_attention.query.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_16.self_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_2.self_attention.query.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_10.attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_10.attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_17.mlp.wi_0.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_17.mlp.wi_0.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_17.self_attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_5.attention.out.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_9.attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_9.attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.temporal_decoder.layers_17.self_attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_16.attention.key.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_2.self_attention.value.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_0.attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.decoder.decoder.depth_decoder.depth_layers_2.self_attention.value.kernel/0.1 filter=lfs diff=lfs merge=lfs -text +checkpoints/llm_large_x3047_c1860k/target.encoder.layers_16.attention.key.kernel/0.0 filter=lfs diff=lfs merge=lfs -text diff --git a/checkpoints/llm_base_x4286_c1860k/state.param_states.decoder.decoder.depth_decoder.depth_layers_1.pre_mlp_layer_norm.scale.v/.zarray b/checkpoints/llm_base_x4286_c1860k/state.param_states.decoder.decoder.depth_decoder.depth_layers_1.pre_mlp_layer_norm.scale.v/.zarray new file mode 100644 index 0000000000000000000000000000000000000000..e95ae447caec523c8222a18a5ec1d3e0d422b339 --- /dev/null +++ b/checkpoints/llm_base_x4286_c1860k/state.param_states.decoder.decoder.depth_decoder.depth_layers_1.pre_mlp_layer_norm.scale.v/.zarray @@ -0,0 +1 @@ +{"chunks":[768],"compressor":{"id":"gzip","level":1},"dimension_separator":".","dtype":"