| { |
| "schema": "trellis-webgpu-pack", |
| "version": 1, |
| "stage": "textEncoder", |
| "upstreamRevision": "32bd64288804d66eefd0ccbe215aa642df71cc41", |
| "shards": [ |
| { |
| "file": "stage-000.bin", |
| "bytes": 118272, |
| "sha256": "c8abf2c72c54162311663abedd016a617af6a0e2f59f30a7f2fa38b866a5d773" |
| }, |
| { |
| "file": "stage-001.bin", |
| "bytes": 75890688, |
| "sha256": "22e014e18663bc94e798325d3db2a2553220bb14a2fdfb869b4b37cf0fd6a12c" |
| }, |
| { |
| "file": "stage-002.bin", |
| "bytes": 65883648, |
| "sha256": "160009eb7364d43618f5c32820762e40a42cfd6fd4d800bd3d46d9ac655b7ae6" |
| }, |
| { |
| "file": "stage-003.bin", |
| "bytes": 21947904, |
| "sha256": "65e8037f74be967ab4e0f45f189e72be3379ccefe92809e537a6443c213fe987" |
| } |
| ], |
| "tensors": { |
| "embeddings.position_embedding.weight": { |
| "shape": [ |
| 77, |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 0, |
| "offset": 0, |
| "bytes": 118272 |
| } |
| }, |
| "embeddings.token_embedding.weight": { |
| "shape": [ |
| 49408, |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 1, |
| "offset": 0, |
| "bytes": 75890688 |
| } |
| }, |
| "encoder.layers.0.layer_norm1.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 0, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.0.layer_norm1.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 1536, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.0.layer_norm2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 3072, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.0.layer_norm2.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 4608, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.0.mlp.fc1.bias": { |
| "shape": [ |
| 3072 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 6144, |
| "bytes": 6144 |
| } |
| }, |
| "encoder.layers.0.mlp.fc1.weight": { |
| "shape": [ |
| 3072, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 3072, |
| "columns": 768, |
| "padded_rows": 3072, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 12288, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 2371584, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.0.mlp.fc2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 2445312, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.0.mlp.fc2.weight": { |
| "shape": [ |
| 768, |
| 3072 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 3072, |
| "padded_rows": 768, |
| "padded_columns": 3072, |
| "data": { |
| "shard": 2, |
| "offset": 2446848, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 4806144, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.0.self_attn.k_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 4879872, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.0.self_attn.k_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 4881408, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 5471232, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.0.self_attn.out_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 5489664, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.0.self_attn.out_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 5491200, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 6081024, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.0.self_attn.q_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 6099456, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.0.self_attn.q_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 6100992, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 6690816, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.0.self_attn.v_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 6709248, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.0.self_attn.v_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 6710784, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 7300608, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.1.layer_norm1.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 7319040, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.1.layer_norm1.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 7320576, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.1.layer_norm2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 7322112, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.1.layer_norm2.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 7323648, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.1.mlp.fc1.bias": { |
| "shape": [ |
| 3072 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 7325184, |
| "bytes": 6144 |
| } |
| }, |
| "encoder.layers.1.mlp.fc1.weight": { |
| "shape": [ |
| 3072, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 3072, |
| "columns": 768, |
| "padded_rows": 3072, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 7331328, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 9690624, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.1.mlp.fc2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 9764352, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.1.mlp.fc2.weight": { |
| "shape": [ |
| 768, |
| 3072 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 3072, |
| "padded_rows": 768, |
| "padded_columns": 3072, |
| "data": { |
| "shard": 2, |
| "offset": 9765888, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 12125184, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.1.self_attn.k_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 12198912, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.1.self_attn.k_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 12200448, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 12790272, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.1.self_attn.out_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 12808704, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.1.self_attn.out_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 12810240, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 13400064, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.1.self_attn.q_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 13418496, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.1.self_attn.q_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 13420032, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 14009856, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.1.self_attn.v_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 14028288, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.1.self_attn.v_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 14029824, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 14619648, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.10.layer_norm1.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 14638080, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.10.layer_norm1.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 14639616, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.10.layer_norm2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 14641152, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.10.layer_norm2.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 14642688, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.10.mlp.fc1.bias": { |
| "shape": [ |
| 3072 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 14644224, |
| "bytes": 6144 |
| } |
| }, |
| "encoder.layers.10.mlp.fc1.weight": { |
| "shape": [ |
| 3072, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 3072, |
| "columns": 768, |
| "padded_rows": 3072, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 14650368, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 17009664, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.10.mlp.fc2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 17083392, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.10.mlp.fc2.weight": { |
| "shape": [ |
| 768, |
| 3072 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 3072, |
| "padded_rows": 768, |
| "padded_columns": 3072, |
| "data": { |
| "shard": 2, |
| "offset": 17084928, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 19444224, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.10.self_attn.k_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 19517952, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.10.self_attn.k_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 19519488, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 20109312, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.10.self_attn.out_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 20127744, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.10.self_attn.out_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 20129280, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 20719104, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.10.self_attn.q_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 20737536, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.10.self_attn.q_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 20739072, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 21328896, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.10.self_attn.v_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 21347328, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.10.self_attn.v_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 21348864, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 21938688, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.11.layer_norm1.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 21957120, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.11.layer_norm1.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 21958656, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.11.layer_norm2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 21960192, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.11.layer_norm2.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 21961728, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.11.mlp.fc1.bias": { |
| "shape": [ |
| 3072 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 21963264, |
| "bytes": 6144 |
| } |
| }, |
| "encoder.layers.11.mlp.fc1.weight": { |
| "shape": [ |
| 3072, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 3072, |
| "columns": 768, |
| "padded_rows": 3072, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 21969408, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 24328704, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.11.mlp.fc2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 24402432, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.11.mlp.fc2.weight": { |
| "shape": [ |
| 768, |
| 3072 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 3072, |
| "padded_rows": 768, |
| "padded_columns": 3072, |
| "data": { |
| "shard": 2, |
| "offset": 24403968, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 26763264, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.11.self_attn.k_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 26836992, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.11.self_attn.k_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 26838528, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 27428352, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.11.self_attn.out_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 27446784, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.11.self_attn.out_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 27448320, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 28038144, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.11.self_attn.q_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 28056576, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.11.self_attn.q_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 28058112, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 28647936, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.11.self_attn.v_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 28666368, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.11.self_attn.v_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 28667904, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 29257728, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.2.layer_norm1.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 29276160, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.2.layer_norm1.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 29277696, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.2.layer_norm2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 29279232, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.2.layer_norm2.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 29280768, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.2.mlp.fc1.bias": { |
| "shape": [ |
| 3072 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 29282304, |
| "bytes": 6144 |
| } |
| }, |
| "encoder.layers.2.mlp.fc1.weight": { |
| "shape": [ |
| 3072, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 3072, |
| "columns": 768, |
| "padded_rows": 3072, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 29288448, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 31647744, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.2.mlp.fc2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 31721472, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.2.mlp.fc2.weight": { |
| "shape": [ |
| 768, |
| 3072 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 3072, |
| "padded_rows": 768, |
| "padded_columns": 3072, |
| "data": { |
| "shard": 2, |
| "offset": 31723008, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 34082304, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.2.self_attn.k_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 34156032, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.2.self_attn.k_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 34157568, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 34747392, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.2.self_attn.out_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 34765824, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.2.self_attn.out_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 34767360, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 35357184, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.2.self_attn.q_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 35375616, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.2.self_attn.q_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 35377152, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 35966976, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.2.self_attn.v_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 35985408, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.2.self_attn.v_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 35986944, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 36576768, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.3.layer_norm1.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 36595200, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.3.layer_norm1.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 36596736, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.3.layer_norm2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 36598272, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.3.layer_norm2.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 36599808, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.3.mlp.fc1.bias": { |
| "shape": [ |
| 3072 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 36601344, |
| "bytes": 6144 |
| } |
| }, |
| "encoder.layers.3.mlp.fc1.weight": { |
| "shape": [ |
| 3072, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 3072, |
| "columns": 768, |
| "padded_rows": 3072, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 36607488, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 38966784, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.3.mlp.fc2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 39040512, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.3.mlp.fc2.weight": { |
| "shape": [ |
| 768, |
| 3072 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 3072, |
| "padded_rows": 768, |
| "padded_columns": 3072, |
| "data": { |
| "shard": 2, |
| "offset": 39042048, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 41401344, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.3.self_attn.k_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 41475072, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.3.self_attn.k_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 41476608, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 42066432, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.3.self_attn.out_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 42084864, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.3.self_attn.out_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 42086400, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 42676224, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.3.self_attn.q_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 42694656, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.3.self_attn.q_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 42696192, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 43286016, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.3.self_attn.v_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 43304448, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.3.self_attn.v_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 43305984, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 43895808, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.4.layer_norm1.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 43914240, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.4.layer_norm1.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 43915776, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.4.layer_norm2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 43917312, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.4.layer_norm2.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 43918848, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.4.mlp.fc1.bias": { |
| "shape": [ |
| 3072 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 43920384, |
| "bytes": 6144 |
| } |
| }, |
| "encoder.layers.4.mlp.fc1.weight": { |
| "shape": [ |
| 3072, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 3072, |
| "columns": 768, |
| "padded_rows": 3072, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 43926528, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 46285824, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.4.mlp.fc2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 46359552, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.4.mlp.fc2.weight": { |
| "shape": [ |
| 768, |
| 3072 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 3072, |
| "padded_rows": 768, |
| "padded_columns": 3072, |
| "data": { |
| "shard": 2, |
| "offset": 46361088, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 48720384, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.4.self_attn.k_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 48794112, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.4.self_attn.k_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 48795648, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 49385472, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.4.self_attn.out_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 49403904, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.4.self_attn.out_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 49405440, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 49995264, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.4.self_attn.q_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 50013696, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.4.self_attn.q_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 50015232, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 50605056, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.4.self_attn.v_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 50623488, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.4.self_attn.v_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 50625024, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 51214848, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.5.layer_norm1.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 51233280, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.5.layer_norm1.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 51234816, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.5.layer_norm2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 51236352, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.5.layer_norm2.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 51237888, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.5.mlp.fc1.bias": { |
| "shape": [ |
| 3072 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 51239424, |
| "bytes": 6144 |
| } |
| }, |
| "encoder.layers.5.mlp.fc1.weight": { |
| "shape": [ |
| 3072, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 3072, |
| "columns": 768, |
| "padded_rows": 3072, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 51245568, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 53604864, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.5.mlp.fc2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 53678592, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.5.mlp.fc2.weight": { |
| "shape": [ |
| 768, |
| 3072 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 3072, |
| "padded_rows": 768, |
| "padded_columns": 3072, |
| "data": { |
| "shard": 2, |
| "offset": 53680128, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 56039424, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.5.self_attn.k_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 56113152, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.5.self_attn.k_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 56114688, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 56704512, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.5.self_attn.out_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 56722944, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.5.self_attn.out_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 56724480, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 57314304, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.5.self_attn.q_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 57332736, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.5.self_attn.q_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 57334272, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 57924096, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.5.self_attn.v_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 57942528, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.5.self_attn.v_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 57944064, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 58533888, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.6.layer_norm1.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 58552320, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.6.layer_norm1.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 58553856, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.6.layer_norm2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 58555392, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.6.layer_norm2.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 58556928, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.6.mlp.fc1.bias": { |
| "shape": [ |
| 3072 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 58558464, |
| "bytes": 6144 |
| } |
| }, |
| "encoder.layers.6.mlp.fc1.weight": { |
| "shape": [ |
| 3072, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 3072, |
| "columns": 768, |
| "padded_rows": 3072, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 58564608, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 60923904, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.6.mlp.fc2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 60997632, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.6.mlp.fc2.weight": { |
| "shape": [ |
| 768, |
| 3072 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 3072, |
| "padded_rows": 768, |
| "padded_columns": 3072, |
| "data": { |
| "shard": 2, |
| "offset": 60999168, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 63358464, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.6.self_attn.k_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 63432192, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.6.self_attn.k_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 63433728, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 64023552, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.6.self_attn.out_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 64041984, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.6.self_attn.out_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 64043520, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 64633344, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.6.self_attn.q_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 64651776, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.6.self_attn.q_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 64653312, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 65243136, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.6.self_attn.v_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 65261568, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.6.self_attn.v_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 2, |
| "offset": 65263104, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 2, |
| "offset": 65852928, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.7.layer_norm1.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 65871360, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.7.layer_norm1.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 65872896, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.7.layer_norm2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 65874432, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.7.layer_norm2.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 65875968, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.7.mlp.fc1.bias": { |
| "shape": [ |
| 3072 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 2, |
| "offset": 65877504, |
| "bytes": 6144 |
| } |
| }, |
| "encoder.layers.7.mlp.fc1.weight": { |
| "shape": [ |
| 3072, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 3072, |
| "columns": 768, |
| "padded_rows": 3072, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 0, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 2359296, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.7.mlp.fc2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 2433024, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.7.mlp.fc2.weight": { |
| "shape": [ |
| 768, |
| 3072 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 3072, |
| "padded_rows": 768, |
| "padded_columns": 3072, |
| "data": { |
| "shard": 3, |
| "offset": 2434560, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 4793856, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.7.self_attn.k_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 4867584, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.7.self_attn.k_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 4869120, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 5458944, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.7.self_attn.out_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 5477376, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.7.self_attn.out_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 5478912, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 6068736, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.7.self_attn.q_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 6087168, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.7.self_attn.q_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 6088704, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 6678528, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.7.self_attn.v_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 6696960, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.7.self_attn.v_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 6698496, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 7288320, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.8.layer_norm1.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 7306752, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.8.layer_norm1.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 7308288, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.8.layer_norm2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 7309824, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.8.layer_norm2.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 7311360, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.8.mlp.fc1.bias": { |
| "shape": [ |
| 3072 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 7312896, |
| "bytes": 6144 |
| } |
| }, |
| "encoder.layers.8.mlp.fc1.weight": { |
| "shape": [ |
| 3072, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 3072, |
| "columns": 768, |
| "padded_rows": 3072, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 7319040, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 9678336, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.8.mlp.fc2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 9752064, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.8.mlp.fc2.weight": { |
| "shape": [ |
| 768, |
| 3072 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 3072, |
| "padded_rows": 768, |
| "padded_columns": 3072, |
| "data": { |
| "shard": 3, |
| "offset": 9753600, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 12112896, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.8.self_attn.k_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 12186624, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.8.self_attn.k_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 12188160, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 12777984, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.8.self_attn.out_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 12796416, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.8.self_attn.out_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 12797952, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 13387776, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.8.self_attn.q_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 13406208, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.8.self_attn.q_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 13407744, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 13997568, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.8.self_attn.v_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 14016000, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.8.self_attn.v_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 14017536, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 14607360, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.9.layer_norm1.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 14625792, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.9.layer_norm1.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 14627328, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.9.layer_norm2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 14628864, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.9.layer_norm2.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 14630400, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.9.mlp.fc1.bias": { |
| "shape": [ |
| 3072 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 14631936, |
| "bytes": 6144 |
| } |
| }, |
| "encoder.layers.9.mlp.fc1.weight": { |
| "shape": [ |
| 3072, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 3072, |
| "columns": 768, |
| "padded_rows": 3072, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 14638080, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 16997376, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.9.mlp.fc2.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 17071104, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.9.mlp.fc2.weight": { |
| "shape": [ |
| 768, |
| 3072 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 3072, |
| "padded_rows": 768, |
| "padded_columns": 3072, |
| "data": { |
| "shard": 3, |
| "offset": 17072640, |
| "bytes": 2359296 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 19431936, |
| "bytes": 73728 |
| } |
| }, |
| "encoder.layers.9.self_attn.k_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 19505664, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.9.self_attn.k_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 19507200, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 20097024, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.9.self_attn.out_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 20115456, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.9.self_attn.out_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 20116992, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 20706816, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.9.self_attn.q_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 20725248, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.9.self_attn.q_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 20726784, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 21316608, |
| "bytes": 18432 |
| } |
| }, |
| "encoder.layers.9.self_attn.v_proj.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 21335040, |
| "bytes": 1536 |
| } |
| }, |
| "encoder.layers.9.self_attn.v_proj.weight": { |
| "shape": [ |
| 768, |
| 768 |
| ], |
| "encoding": "symmetric-quantized", |
| "bits": 8, |
| "group_size": 64, |
| "output_tile": 16, |
| "rows": 768, |
| "columns": 768, |
| "padded_rows": 768, |
| "padded_columns": 768, |
| "data": { |
| "shard": 3, |
| "offset": 21336576, |
| "bytes": 589824 |
| }, |
| "scales": { |
| "shard": 3, |
| "offset": 21926400, |
| "bytes": 18432 |
| } |
| }, |
| "final_layer_norm.bias": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 21944832, |
| "bytes": 1536 |
| } |
| }, |
| "final_layer_norm.weight": { |
| "shape": [ |
| 768 |
| ], |
| "encoding": "raw", |
| "dtype": "f16", |
| "data": { |
| "shard": 3, |
| "offset": 21946368, |
| "bytes": 1536 |
| } |
| } |
| } |
| } |
|
|