Download gaussianDecoder/manifest.json from needle-tools/LiTo-webgpu: direct link, hf CLI and curl.
- Browser
- Download file 85.6 kB
-
https://huggingface.co/needle-tools/LiTo-webgpu/resolve/main/gaussianDecoder/manifest.json
- Command line
-
hf download hf://needle-tools/LiTo-webgpu/gaussianDecoder/manifest.json
-
curl -L -o manifest.json https://huggingface.co/needle-tools/LiTo-webgpu/resolve/main/gaussianDecoder/manifest.json
85.6 kB
| { | |
| "schema": "trellis-webgpu-pack", | |
| "version": 1, | |
| "stage": "litoGaussianDecoder", | |
| "upstreamRevision": "apple/ml-lito", | |
| "shards": [ | |
| { | |
| "file": "stage-000.bin", | |
| "bytes": 65748480, | |
| "sha256": "5de9eb5c449bded4954389baf8e15b17df3af859ad753bc0fa5a4c3862051d84" | |
| }, | |
| { | |
| "file": "stage-001.bin", | |
| "bytes": 34025536, | |
| "sha256": "1a0c141a891da9e11568d8acd47fd6a85a7b8e6c18b2899359f3fc290705c3bc" | |
| } | |
| ], | |
| "tensors": { | |
| "gs_output_color_mlp.final_layer.linear.bias": { | |
| "shape": [ | |
| 3136 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 0, | |
| "bytes": 6272 | |
| } | |
| }, | |
| "gs_output_color_mlp.final_layer.linear.weight": { | |
| "shape": [ | |
| 3136, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 6400, | |
| "bytes": 3211264 | |
| } | |
| }, | |
| "gs_output_color_mlp.final_layer.norm_final.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 3217664, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "gs_output_color_mlp.final_layer.norm_final.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 3218688, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "gs_output_color_mlp.mlps.0.w1.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 3219712, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "gs_output_color_mlp.mlps.0.w2.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 3744000, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "gs_output_color_mlp.mlps.0.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 4268288, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "gs_output_color_mlp.norms.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 4792576, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "gs_output_color_mlp.norms.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 4793600, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "gs_output_shape_mlp.final_layer.linear.bias": { | |
| "shape": [ | |
| 640 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 4794624, | |
| "bytes": 1280 | |
| } | |
| }, | |
| "gs_output_shape_mlp.final_layer.linear.weight": { | |
| "shape": [ | |
| 640, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 4795904, | |
| "bytes": 655360 | |
| } | |
| }, | |
| "gs_output_shape_mlp.final_layer.norm_final.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 5451264, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "gs_output_shape_mlp.final_layer.norm_final.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 5452288, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "gs_output_shape_mlp.mlps.0.w1.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 5453312, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "gs_output_shape_mlp.mlps.0.w2.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 5977600, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "gs_output_shape_mlp.mlps.0.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 6501888, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "gs_output_shape_mlp.norms.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 7026176, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "gs_output_shape_mlp.norms.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 7027200, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_layer.layernorm_kv.bias": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 7028224, | |
| "bytes": 64 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_layer.layernorm_kv.weight": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 7028480, | |
| "bytes": 64 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_layer.layernorm_q.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 7028736, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_layer.layernorm_q.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 7029760, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_layer.linear_kv.bias": { | |
| "shape": [ | |
| 1024 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 7030784, | |
| "bytes": 2048 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_layer.linear_kv.weight": { | |
| "shape": [ | |
| 1024, | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 7032832, | |
| "bytes": 65536 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_layer.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 7098368, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_layer.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 7099392, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_layer.linear_q.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 7623680, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_layer.linear_q.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 7624704, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_layer.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 8148992, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_layer.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 8150016, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_ln.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 8151040, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_ln.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 8152064, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_mlp.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 8153088, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 9201664, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_mlp.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 9234432, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 10283008, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.0.ca_mlp.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 0, | |
| "offset": 10315776, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 11364352, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.0.kv_linear.weight": { | |
| "shape": [ | |
| 32, | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 11397120, | |
| "bytes": 2048 | |
| } | |
| }, | |
| "perceiver.blocks.0.ln1_layers.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 11399168, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ln1_layers.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 11400192, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ln1_layers.1.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 11401216, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ln1_layers.1.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 11402240, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ln2_layers.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 11403264, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ln2_layers.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 11404288, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ln2_layers.1.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 11405312, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.ln2_layers.1.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 11406336, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.mlp_layers.0.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 11407360, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 12455936, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.0.mlp_layers.0.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 12488704, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 13537280, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.0.mlp_layers.0.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 0, | |
| "offset": 13570048, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 14618624, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.0.mlp_layers.1.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 14651392, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 15699968, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.0.mlp_layers.1.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 15732736, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 16781312, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.0.mlp_layers.1.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 0, | |
| "offset": 16814080, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 17862656, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.0.sa_layers.0.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 17895424, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.sa_layers.0.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 17896448, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.0.sa_layers.0.linear_qkv.bias": { | |
| "shape": [ | |
| 1536 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 18420736, | |
| "bytes": 3072 | |
| } | |
| }, | |
| "perceiver.blocks.0.sa_layers.0.linear_qkv.weight": { | |
| "shape": [ | |
| 1536, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 18423808, | |
| "bytes": 1572864 | |
| } | |
| }, | |
| "perceiver.blocks.0.sa_layers.0.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 19996672, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.sa_layers.0.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 19997696, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.sa_layers.1.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 19998720, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.sa_layers.1.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 19999744, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.0.sa_layers.1.linear_qkv.bias": { | |
| "shape": [ | |
| 1536 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 20524032, | |
| "bytes": 3072 | |
| } | |
| }, | |
| "perceiver.blocks.0.sa_layers.1.linear_qkv.weight": { | |
| "shape": [ | |
| 1536, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 20527104, | |
| "bytes": 1572864 | |
| } | |
| }, | |
| "perceiver.blocks.0.sa_layers.1.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 22099968, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.0.sa_layers.1.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 22100992, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_layer.layernorm_kv.bias": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 22102016, | |
| "bytes": 64 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_layer.layernorm_kv.weight": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 22102272, | |
| "bytes": 64 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_layer.layernorm_q.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 22102528, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_layer.layernorm_q.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 22103552, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_layer.linear_kv.bias": { | |
| "shape": [ | |
| 1024 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 22104576, | |
| "bytes": 2048 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_layer.linear_kv.weight": { | |
| "shape": [ | |
| 1024, | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 22106624, | |
| "bytes": 65536 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_layer.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 22172160, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_layer.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 22173184, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_layer.linear_q.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 22697472, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_layer.linear_q.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 22698496, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_layer.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 23222784, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_layer.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 23223808, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_ln.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 23224832, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_ln.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 23225856, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_mlp.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 23226880, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 24275456, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_mlp.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 24308224, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 25356800, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.1.ca_mlp.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 0, | |
| "offset": 25389568, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 26438144, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.1.kv_linear.weight": { | |
| "shape": [ | |
| 32, | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 26470912, | |
| "bytes": 2048 | |
| } | |
| }, | |
| "perceiver.blocks.1.ln1_layers.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 26472960, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ln1_layers.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 26473984, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ln1_layers.1.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 26475008, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ln1_layers.1.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 26476032, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ln2_layers.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 26477056, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ln2_layers.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 26478080, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ln2_layers.1.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 26479104, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.ln2_layers.1.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 26480128, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.mlp_layers.0.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 26481152, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 27529728, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.1.mlp_layers.0.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 27562496, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 28611072, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.1.mlp_layers.0.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 0, | |
| "offset": 28643840, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 29692416, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.1.mlp_layers.1.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 29725184, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 30773760, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.1.mlp_layers.1.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 30806528, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 31855104, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.1.mlp_layers.1.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 0, | |
| "offset": 31887872, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 32936448, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.1.sa_layers.0.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 32969216, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.sa_layers.0.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 32970240, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.1.sa_layers.0.linear_qkv.bias": { | |
| "shape": [ | |
| 1536 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 33494528, | |
| "bytes": 3072 | |
| } | |
| }, | |
| "perceiver.blocks.1.sa_layers.0.linear_qkv.weight": { | |
| "shape": [ | |
| 1536, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 33497600, | |
| "bytes": 1572864 | |
| } | |
| }, | |
| "perceiver.blocks.1.sa_layers.0.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 35070464, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.sa_layers.0.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 35071488, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.sa_layers.1.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 35072512, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.sa_layers.1.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 35073536, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.1.sa_layers.1.linear_qkv.bias": { | |
| "shape": [ | |
| 1536 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 35597824, | |
| "bytes": 3072 | |
| } | |
| }, | |
| "perceiver.blocks.1.sa_layers.1.linear_qkv.weight": { | |
| "shape": [ | |
| 1536, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 35600896, | |
| "bytes": 1572864 | |
| } | |
| }, | |
| "perceiver.blocks.1.sa_layers.1.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 37173760, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.1.sa_layers.1.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 37174784, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_layer.layernorm_kv.bias": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 37175808, | |
| "bytes": 64 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_layer.layernorm_kv.weight": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 37176064, | |
| "bytes": 64 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_layer.layernorm_q.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 37176320, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_layer.layernorm_q.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 37177344, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_layer.linear_kv.bias": { | |
| "shape": [ | |
| 1024 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 37178368, | |
| "bytes": 2048 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_layer.linear_kv.weight": { | |
| "shape": [ | |
| 1024, | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 37180416, | |
| "bytes": 65536 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_layer.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 37245952, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_layer.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 37246976, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_layer.linear_q.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 37771264, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_layer.linear_q.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 37772288, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_layer.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 38296576, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_layer.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 38297600, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_ln.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 38298624, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_ln.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 38299648, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_mlp.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 38300672, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 39349248, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_mlp.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 39382016, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 40430592, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.2.ca_mlp.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 0, | |
| "offset": 40463360, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 41511936, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.2.kv_linear.weight": { | |
| "shape": [ | |
| 32, | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 41544704, | |
| "bytes": 2048 | |
| } | |
| }, | |
| "perceiver.blocks.2.ln1_layers.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 41546752, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ln1_layers.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 41547776, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ln1_layers.1.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 41548800, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ln1_layers.1.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 41549824, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ln2_layers.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 41550848, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ln2_layers.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 41551872, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ln2_layers.1.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 41552896, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.ln2_layers.1.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 41553920, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.mlp_layers.0.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 41554944, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 42603520, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.2.mlp_layers.0.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 42636288, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 43684864, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.2.mlp_layers.0.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 0, | |
| "offset": 43717632, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 44766208, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.2.mlp_layers.1.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 44798976, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 45847552, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.2.mlp_layers.1.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 45880320, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 46928896, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.2.mlp_layers.1.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 0, | |
| "offset": 46961664, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 48010240, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.2.sa_layers.0.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 48043008, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.sa_layers.0.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 48044032, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.2.sa_layers.0.linear_qkv.bias": { | |
| "shape": [ | |
| 1536 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 48568320, | |
| "bytes": 3072 | |
| } | |
| }, | |
| "perceiver.blocks.2.sa_layers.0.linear_qkv.weight": { | |
| "shape": [ | |
| 1536, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 48571392, | |
| "bytes": 1572864 | |
| } | |
| }, | |
| "perceiver.blocks.2.sa_layers.0.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 50144256, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.sa_layers.0.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 50145280, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.sa_layers.1.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 50146304, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.sa_layers.1.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 50147328, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.2.sa_layers.1.linear_qkv.bias": { | |
| "shape": [ | |
| 1536 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 50671616, | |
| "bytes": 3072 | |
| } | |
| }, | |
| "perceiver.blocks.2.sa_layers.1.linear_qkv.weight": { | |
| "shape": [ | |
| 1536, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 50674688, | |
| "bytes": 1572864 | |
| } | |
| }, | |
| "perceiver.blocks.2.sa_layers.1.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 52247552, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.2.sa_layers.1.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 52248576, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_layer.layernorm_kv.bias": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 52249600, | |
| "bytes": 64 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_layer.layernorm_kv.weight": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 52249856, | |
| "bytes": 64 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_layer.layernorm_q.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 52250112, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_layer.layernorm_q.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 52251136, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_layer.linear_kv.bias": { | |
| "shape": [ | |
| 1024 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 52252160, | |
| "bytes": 2048 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_layer.linear_kv.weight": { | |
| "shape": [ | |
| 1024, | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 52254208, | |
| "bytes": 65536 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_layer.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 52319744, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_layer.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 52320768, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_layer.linear_q.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 52845056, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_layer.linear_q.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 52846080, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_layer.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 53370368, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_layer.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 53371392, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_ln.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 53372416, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_ln.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 53373440, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_mlp.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 53374464, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 54423040, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_mlp.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 54455808, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 55504384, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.3.ca_mlp.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 0, | |
| "offset": 55537152, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 56585728, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.3.kv_linear.weight": { | |
| "shape": [ | |
| 32, | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 56618496, | |
| "bytes": 2048 | |
| } | |
| }, | |
| "perceiver.blocks.3.ln1_layers.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 56620544, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ln1_layers.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 56621568, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ln1_layers.1.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 56622592, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ln1_layers.1.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 56623616, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ln2_layers.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 56624640, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ln2_layers.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 56625664, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ln2_layers.1.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 56626688, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.ln2_layers.1.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 56627712, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.mlp_layers.0.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 56628736, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 57677312, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.3.mlp_layers.0.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 57710080, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 58758656, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.3.mlp_layers.0.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 0, | |
| "offset": 58791424, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 59840000, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.3.mlp_layers.1.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 59872768, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 60921344, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.3.mlp_layers.1.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 0, | |
| "offset": 60954112, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 62002688, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.3.mlp_layers.1.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 0, | |
| "offset": 62035456, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 0, | |
| "offset": 63084032, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.3.sa_layers.0.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 63116800, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.sa_layers.0.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 63117824, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.3.sa_layers.0.linear_qkv.bias": { | |
| "shape": [ | |
| 1536 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 63642112, | |
| "bytes": 3072 | |
| } | |
| }, | |
| "perceiver.blocks.3.sa_layers.0.linear_qkv.weight": { | |
| "shape": [ | |
| 1536, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 63645184, | |
| "bytes": 1572864 | |
| } | |
| }, | |
| "perceiver.blocks.3.sa_layers.0.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 65218048, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.sa_layers.0.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 65219072, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.sa_layers.1.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 65220096, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.sa_layers.1.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 65221120, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.3.sa_layers.1.linear_qkv.bias": { | |
| "shape": [ | |
| 1536 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 0, | |
| "offset": 65745408, | |
| "bytes": 3072 | |
| } | |
| }, | |
| "perceiver.blocks.3.sa_layers.1.linear_qkv.weight": { | |
| "shape": [ | |
| 1536, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 0, | |
| "bytes": 1572864 | |
| } | |
| }, | |
| "perceiver.blocks.3.sa_layers.1.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 1572864, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.3.sa_layers.1.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 1573888, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_layer.layernorm_kv.bias": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 1574912, | |
| "bytes": 64 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_layer.layernorm_kv.weight": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 1575168, | |
| "bytes": 64 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_layer.layernorm_q.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 1575424, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_layer.layernorm_q.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 1576448, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_layer.linear_kv.bias": { | |
| "shape": [ | |
| 1024 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 1577472, | |
| "bytes": 2048 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_layer.linear_kv.weight": { | |
| "shape": [ | |
| 1024, | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 1579520, | |
| "bytes": 65536 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_layer.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 1645056, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_layer.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 1646080, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_layer.linear_q.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 2170368, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_layer.linear_q.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 2171392, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_layer.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 2695680, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_layer.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 2696704, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_ln.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 2697728, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_ln.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 2698752, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_mlp.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 1, | |
| "offset": 2699776, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 3748352, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_mlp.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 1, | |
| "offset": 3781120, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 4829696, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.4.ca_mlp.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 1, | |
| "offset": 4862464, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 5911040, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.4.kv_linear.weight": { | |
| "shape": [ | |
| 32, | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 5943808, | |
| "bytes": 2048 | |
| } | |
| }, | |
| "perceiver.blocks.4.ln1_layers.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 5945856, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ln1_layers.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 5946880, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ln1_layers.1.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 5947904, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ln1_layers.1.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 5948928, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ln2_layers.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 5949952, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ln2_layers.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 5950976, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ln2_layers.1.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 5952000, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.ln2_layers.1.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 5953024, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.mlp_layers.0.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 1, | |
| "offset": 5954048, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 7002624, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.4.mlp_layers.0.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 1, | |
| "offset": 7035392, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 8083968, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.4.mlp_layers.0.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 1, | |
| "offset": 8116736, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 9165312, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.4.mlp_layers.1.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 1, | |
| "offset": 9198080, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 10246656, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.4.mlp_layers.1.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 1, | |
| "offset": 10279424, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 11328000, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.4.mlp_layers.1.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 1, | |
| "offset": 11360768, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 12409344, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.4.sa_layers.0.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 12442112, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.sa_layers.0.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 12443136, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.4.sa_layers.0.linear_qkv.bias": { | |
| "shape": [ | |
| 1536 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 12967424, | |
| "bytes": 3072 | |
| } | |
| }, | |
| "perceiver.blocks.4.sa_layers.0.linear_qkv.weight": { | |
| "shape": [ | |
| 1536, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 12970496, | |
| "bytes": 1572864 | |
| } | |
| }, | |
| "perceiver.blocks.4.sa_layers.0.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 14543360, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.sa_layers.0.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 14544384, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.sa_layers.1.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 14545408, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.sa_layers.1.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 14546432, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.4.sa_layers.1.linear_qkv.bias": { | |
| "shape": [ | |
| 1536 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 15070720, | |
| "bytes": 3072 | |
| } | |
| }, | |
| "perceiver.blocks.4.sa_layers.1.linear_qkv.weight": { | |
| "shape": [ | |
| 1536, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 15073792, | |
| "bytes": 1572864 | |
| } | |
| }, | |
| "perceiver.blocks.4.sa_layers.1.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 16646656, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.4.sa_layers.1.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 16647680, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_layer.layernorm_kv.bias": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 16648704, | |
| "bytes": 64 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_layer.layernorm_kv.weight": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 16648960, | |
| "bytes": 64 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_layer.layernorm_q.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 16649216, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_layer.layernorm_q.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 16650240, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_layer.linear_kv.bias": { | |
| "shape": [ | |
| 1024 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 16651264, | |
| "bytes": 2048 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_layer.linear_kv.weight": { | |
| "shape": [ | |
| 1024, | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 16653312, | |
| "bytes": 65536 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_layer.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 16718848, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_layer.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 16719872, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_layer.linear_q.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 17244160, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_layer.linear_q.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 17245184, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_layer.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 17769472, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_layer.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 17770496, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_ln.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 17771520, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_ln.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 17772544, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_mlp.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 1, | |
| "offset": 17773568, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 18822144, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_mlp.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 1, | |
| "offset": 18854912, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 19903488, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.5.ca_mlp.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 1, | |
| "offset": 19936256, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 20984832, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.5.kv_linear.weight": { | |
| "shape": [ | |
| 32, | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 21017600, | |
| "bytes": 2048 | |
| } | |
| }, | |
| "perceiver.blocks.5.ln1_layers.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 21019648, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ln1_layers.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 21020672, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ln1_layers.1.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 21021696, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ln1_layers.1.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 21022720, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ln2_layers.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 21023744, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ln2_layers.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 21024768, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ln2_layers.1.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 21025792, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.ln2_layers.1.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 21026816, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.mlp_layers.0.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 1, | |
| "offset": 21027840, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 22076416, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.5.mlp_layers.0.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 1, | |
| "offset": 22109184, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 23157760, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.5.mlp_layers.0.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 1, | |
| "offset": 23190528, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 24239104, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.5.mlp_layers.1.w1.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 1, | |
| "offset": 24271872, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 25320448, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.5.mlp_layers.1.w2.weight": { | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 2048, | |
| "columns": 512, | |
| "padded_rows": 2048, | |
| "padded_columns": 512, | |
| "data": { | |
| "shard": 1, | |
| "offset": 25353216, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 26401792, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.5.mlp_layers.1.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 2048 | |
| ], | |
| "encoding": "symmetric-quantized", | |
| "bits": 8, | |
| "group_size": 64, | |
| "output_tile": 16, | |
| "rows": 512, | |
| "columns": 2048, | |
| "padded_rows": 512, | |
| "padded_columns": 2048, | |
| "data": { | |
| "shard": 1, | |
| "offset": 26434560, | |
| "bytes": 1048576 | |
| }, | |
| "scales": { | |
| "shard": 1, | |
| "offset": 27483136, | |
| "bytes": 32768 | |
| } | |
| }, | |
| "perceiver.blocks.5.sa_layers.0.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 27515904, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.sa_layers.0.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 27516928, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.5.sa_layers.0.linear_qkv.bias": { | |
| "shape": [ | |
| 1536 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 28041216, | |
| "bytes": 3072 | |
| } | |
| }, | |
| "perceiver.blocks.5.sa_layers.0.linear_qkv.weight": { | |
| "shape": [ | |
| 1536, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 28044288, | |
| "bytes": 1572864 | |
| } | |
| }, | |
| "perceiver.blocks.5.sa_layers.0.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 29617152, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.sa_layers.0.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 29618176, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.sa_layers.1.linear_out.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 29619200, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.sa_layers.1.linear_out.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 29620224, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "perceiver.blocks.5.sa_layers.1.linear_qkv.bias": { | |
| "shape": [ | |
| 1536 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 30144512, | |
| "bytes": 3072 | |
| } | |
| }, | |
| "perceiver.blocks.5.sa_layers.1.linear_qkv.weight": { | |
| "shape": [ | |
| 1536, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 30147584, | |
| "bytes": 1572864 | |
| } | |
| }, | |
| "perceiver.blocks.5.sa_layers.1.rmsnorm_k.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 31720448, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "perceiver.blocks.5.sa_layers.1.rmsnorm_q.scale": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 31721472, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "point_linear.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 31722496, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "point_linear.weight": { | |
| "shape": [ | |
| 512, | |
| 195 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 31723520, | |
| "bytes": 199680 | |
| } | |
| }, | |
| "point_mlp.final_layer.linear.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 31923200, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "point_mlp.final_layer.linear.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 31924224, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "point_mlp.final_layer.norm_final.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 32448512, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "point_mlp.final_layer.norm_final.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 32449536, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "point_mlp.mlps.0.w1.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 32450560, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "point_mlp.mlps.0.w2.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 32974848, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "point_mlp.mlps.0.w3.weight": { | |
| "shape": [ | |
| 512, | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 33499136, | |
| "bytes": 524288 | |
| } | |
| }, | |
| "point_mlp.norms.0.bias": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 34023424, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "point_mlp.norms.0.weight": { | |
| "shape": [ | |
| 512 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 34024448, | |
| "bytes": 1024 | |
| } | |
| }, | |
| "xyz_encoding.freq_bands": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "encoding": "raw", | |
| "dtype": "f16", | |
| "data": { | |
| "shard": 1, | |
| "offset": 34025472, | |
| "bytes": 64 | |
| } | |
| } | |
| } | |
| } | |