| { | |
| "metadata": { | |
| "ParamSize": 461, | |
| "ParamBytes": 1462372080.0, | |
| "BitsPerParam": 4.503512067376192 | |
| }, | |
| "records": [ | |
| { | |
| "dataPath": "params_shard_0.bin", | |
| "format": "raw-shard", | |
| "nbytes": 262145024, | |
| "records": [ | |
| { | |
| "name": "language_model.model.embed_tokens.q_weight", | |
| "shape": [ | |
| 256001, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262145024, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "be106bcbbc90a5b522531270fce4c13c" | |
| }, | |
| { | |
| "dataPath": "params_shard_1.bin", | |
| "format": "raw-shard", | |
| "nbytes": 16777216, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.0.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "8ba40bd8094027119107d904f7ec9eeb" | |
| }, | |
| { | |
| "dataPath": "params_shard_2.bin", | |
| "format": "raw-shard", | |
| "nbytes": 32772224, | |
| "records": [ | |
| { | |
| "name": "language_model.model.embed_tokens.q_scale", | |
| "shape": [ | |
| 256001, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 32768128, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.0.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 32768128 | |
| } | |
| ], | |
| "md5sum": "7d086a8bfd46449366a41e9a3a40ffc3" | |
| }, | |
| { | |
| "dataPath": "params_shard_3.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.0.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "8b0fca162b549f7ea2578cf08ed35e57" | |
| }, | |
| { | |
| "dataPath": "params_shard_4.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.1.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "954599139d618bbfc4a77aacd8a5a529" | |
| }, | |
| { | |
| "dataPath": "params_shard_5.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30482432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.0.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.0.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 2097152 | |
| }, | |
| { | |
| "name": "language_model.model.layers.0.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "language_model.model.layers.0.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 6295552 | |
| }, | |
| { | |
| "name": "language_model.model.layers.0.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 8916992 | |
| }, | |
| { | |
| "name": "language_model.model.layers.0.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 9244672 | |
| }, | |
| { | |
| "name": "language_model.model.layers.0.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 11341824 | |
| }, | |
| { | |
| "name": "language_model.model.layers.1.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 11603968 | |
| }, | |
| { | |
| "name": "language_model.model.layers.1.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 11608064 | |
| }, | |
| { | |
| "name": "language_model.model.layers.1.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 28385280 | |
| } | |
| ], | |
| "md5sum": "692cc3000ce9a28625a249b6cdfbcda0" | |
| }, | |
| { | |
| "dataPath": "params_shard_6.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.10.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "49ba4b5f824c47741c3755cdcad9a315" | |
| }, | |
| { | |
| "dataPath": "params_shard_7.bin", | |
| "format": "raw-shard", | |
| "nbytes": 32583680, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.1.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.1.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 4194304 | |
| }, | |
| { | |
| "name": "language_model.model.layers.1.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 4198400 | |
| }, | |
| { | |
| "name": "language_model.model.layers.1.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 6819840 | |
| }, | |
| { | |
| "name": "language_model.model.layers.1.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 7147520 | |
| }, | |
| { | |
| "name": "language_model.model.layers.1.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 9244672 | |
| }, | |
| { | |
| "name": "language_model.model.layers.10.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 9506816 | |
| }, | |
| { | |
| "name": "language_model.model.layers.10.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 9510912 | |
| }, | |
| { | |
| "name": "language_model.model.layers.10.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 26288128 | |
| }, | |
| { | |
| "name": "language_model.model.layers.10.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 28385280 | |
| }, | |
| { | |
| "name": "language_model.model.layers.10.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 32579584 | |
| } | |
| ], | |
| "md5sum": "f318d6f6099475a48eb2f3ef6464d2d6" | |
| }, | |
| { | |
| "dataPath": "params_shard_8.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.11.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "09a9fd809fc85836eba7fb51d8383569" | |
| }, | |
| { | |
| "dataPath": "params_shard_9.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33431552, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.10.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.10.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 2621440 | |
| }, | |
| { | |
| "name": "language_model.model.layers.10.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 2949120 | |
| }, | |
| { | |
| "name": "language_model.model.layers.10.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 5046272 | |
| }, | |
| { | |
| "name": "language_model.model.layers.11.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 5308416 | |
| }, | |
| { | |
| "name": "language_model.model.layers.11.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 5312512 | |
| }, | |
| { | |
| "name": "language_model.model.layers.11.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 22089728 | |
| }, | |
| { | |
| "name": "language_model.model.layers.11.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 24186880 | |
| }, | |
| { | |
| "name": "language_model.model.layers.11.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 28381184 | |
| }, | |
| { | |
| "name": "language_model.model.layers.11.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 28385280 | |
| }, | |
| { | |
| "name": "language_model.model.layers.11.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 31006720 | |
| }, | |
| { | |
| "name": "language_model.model.layers.11.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 31334400 | |
| } | |
| ], | |
| "md5sum": "c6a3793dafb4d39ff9c1398f5a008f16" | |
| }, | |
| { | |
| "dataPath": "params_shard_10.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.12.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "73fe8073bfcdafcc445a0731c590d654" | |
| }, | |
| { | |
| "dataPath": "params_shard_11.bin", | |
| "format": "raw-shard", | |
| "nbytes": 16777216, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.13.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "8a264753d0cd3a2f8bf3399ab885712d" | |
| }, | |
| { | |
| "dataPath": "params_shard_12.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.13.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "987fac9467c1f3a936bc54f9c20a290f" | |
| }, | |
| { | |
| "dataPath": "params_shard_13.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30748672, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.11.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.12.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 262144 | |
| }, | |
| { | |
| "name": "language_model.model.layers.12.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 266240 | |
| }, | |
| { | |
| "name": "language_model.model.layers.12.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 17043456 | |
| }, | |
| { | |
| "name": "language_model.model.layers.12.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 19140608 | |
| }, | |
| { | |
| "name": "language_model.model.layers.12.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 23334912 | |
| }, | |
| { | |
| "name": "language_model.model.layers.12.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 23339008 | |
| }, | |
| { | |
| "name": "language_model.model.layers.12.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 25960448 | |
| }, | |
| { | |
| "name": "language_model.model.layers.12.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 26288128 | |
| }, | |
| { | |
| "name": "language_model.model.layers.12.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 28385280 | |
| }, | |
| { | |
| "name": "language_model.model.layers.13.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 28647424 | |
| }, | |
| { | |
| "name": "language_model.model.layers.13.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 28651520 | |
| } | |
| ], | |
| "md5sum": "983674c59ed73bcdd87507fba31221b9" | |
| }, | |
| { | |
| "dataPath": "params_shard_14.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.14.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "1966b158adb42928c56153616ae0c89e" | |
| }, | |
| { | |
| "dataPath": "params_shard_15.bin", | |
| "format": "raw-shard", | |
| "nbytes": 32583680, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.13.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.13.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 4194304 | |
| }, | |
| { | |
| "name": "language_model.model.layers.13.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 4198400 | |
| }, | |
| { | |
| "name": "language_model.model.layers.13.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 6819840 | |
| }, | |
| { | |
| "name": "language_model.model.layers.13.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 7147520 | |
| }, | |
| { | |
| "name": "language_model.model.layers.13.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 9244672 | |
| }, | |
| { | |
| "name": "language_model.model.layers.14.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 9506816 | |
| }, | |
| { | |
| "name": "language_model.model.layers.14.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 9510912 | |
| }, | |
| { | |
| "name": "language_model.model.layers.14.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 26288128 | |
| }, | |
| { | |
| "name": "language_model.model.layers.14.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 28385280 | |
| }, | |
| { | |
| "name": "language_model.model.layers.14.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 32579584 | |
| } | |
| ], | |
| "md5sum": "f84c61e739c140fad0a40d168952fbf7" | |
| }, | |
| { | |
| "dataPath": "params_shard_16.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.15.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "bb773466dcdf5aaac3d07ea121808010" | |
| }, | |
| { | |
| "dataPath": "params_shard_17.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33431552, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.14.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.14.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 2621440 | |
| }, | |
| { | |
| "name": "language_model.model.layers.14.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 2949120 | |
| }, | |
| { | |
| "name": "language_model.model.layers.14.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 5046272 | |
| }, | |
| { | |
| "name": "language_model.model.layers.15.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 5308416 | |
| }, | |
| { | |
| "name": "language_model.model.layers.15.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 5312512 | |
| }, | |
| { | |
| "name": "language_model.model.layers.15.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 22089728 | |
| }, | |
| { | |
| "name": "language_model.model.layers.15.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 24186880 | |
| }, | |
| { | |
| "name": "language_model.model.layers.15.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 28381184 | |
| }, | |
| { | |
| "name": "language_model.model.layers.15.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 28385280 | |
| }, | |
| { | |
| "name": "language_model.model.layers.15.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 31006720 | |
| }, | |
| { | |
| "name": "language_model.model.layers.15.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 31334400 | |
| } | |
| ], | |
| "md5sum": "6ae36dfd5c3823cfef45501f84f4fe10" | |
| }, | |
| { | |
| "dataPath": "params_shard_18.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.16.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "4252c4cf07dd00d3f71eaa77818f59b7" | |
| }, | |
| { | |
| "dataPath": "params_shard_19.bin", | |
| "format": "raw-shard", | |
| "nbytes": 16777216, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.2.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "b3faa38ee7028509b1db00910746d1a4" | |
| }, | |
| { | |
| "dataPath": "params_shard_20.bin", | |
| "format": "raw-shard", | |
| "nbytes": 31600640, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.15.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.16.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 262144 | |
| }, | |
| { | |
| "name": "language_model.model.layers.16.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 266240 | |
| }, | |
| { | |
| "name": "language_model.model.layers.16.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 17043456 | |
| }, | |
| { | |
| "name": "language_model.model.layers.16.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 19140608 | |
| }, | |
| { | |
| "name": "language_model.model.layers.16.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 23334912 | |
| }, | |
| { | |
| "name": "language_model.model.layers.16.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 23339008 | |
| }, | |
| { | |
| "name": "language_model.model.layers.16.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 25960448 | |
| }, | |
| { | |
| "name": "language_model.model.layers.16.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 26288128 | |
| }, | |
| { | |
| "name": "language_model.model.layers.16.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 28385280 | |
| }, | |
| { | |
| "name": "language_model.model.layers.17.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 28647424 | |
| }, | |
| { | |
| "name": "language_model.model.layers.17.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 31268864 | |
| }, | |
| { | |
| "name": "language_model.model.layers.2.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 31596544 | |
| } | |
| ], | |
| "md5sum": "7b7f7d6708692ff3b87adc2263d1e558" | |
| }, | |
| { | |
| "dataPath": "params_shard_21.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.2.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "39e015ac9151e7b1f0e7c5e77d8efb7a" | |
| }, | |
| { | |
| "dataPath": "params_shard_22.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.3.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "9b18ee605c969bc73a84168af776781d" | |
| }, | |
| { | |
| "dataPath": "params_shard_23.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30482432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.2.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.2.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 2097152 | |
| }, | |
| { | |
| "name": "language_model.model.layers.2.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 6291456 | |
| }, | |
| { | |
| "name": "language_model.model.layers.2.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 6295552 | |
| }, | |
| { | |
| "name": "language_model.model.layers.2.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 8916992 | |
| }, | |
| { | |
| "name": "language_model.model.layers.2.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 9244672 | |
| }, | |
| { | |
| "name": "language_model.model.layers.2.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 11341824 | |
| }, | |
| { | |
| "name": "language_model.model.layers.3.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 11603968 | |
| }, | |
| { | |
| "name": "language_model.model.layers.3.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 11608064 | |
| }, | |
| { | |
| "name": "language_model.model.layers.3.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 28385280 | |
| } | |
| ], | |
| "md5sum": "2e25a8c93b681166f7f83251f45e16f0" | |
| }, | |
| { | |
| "dataPath": "params_shard_24.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.4.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "0fa6145561de8f87c7b3b3f1469babcf" | |
| }, | |
| { | |
| "dataPath": "params_shard_25.bin", | |
| "format": "raw-shard", | |
| "nbytes": 32583680, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.3.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.3.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 4194304 | |
| }, | |
| { | |
| "name": "language_model.model.layers.3.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 4198400 | |
| }, | |
| { | |
| "name": "language_model.model.layers.3.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 6819840 | |
| }, | |
| { | |
| "name": "language_model.model.layers.3.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 7147520 | |
| }, | |
| { | |
| "name": "language_model.model.layers.3.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 9244672 | |
| }, | |
| { | |
| "name": "language_model.model.layers.4.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 9506816 | |
| }, | |
| { | |
| "name": "language_model.model.layers.4.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 9510912 | |
| }, | |
| { | |
| "name": "language_model.model.layers.4.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 26288128 | |
| }, | |
| { | |
| "name": "language_model.model.layers.4.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 28385280 | |
| }, | |
| { | |
| "name": "language_model.model.layers.4.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 32579584 | |
| } | |
| ], | |
| "md5sum": "a9df183d222564d0d83e9b8ca437ceaf" | |
| }, | |
| { | |
| "dataPath": "params_shard_26.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.5.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "4dda78f0f0c5dc551f0572f703942b94" | |
| }, | |
| { | |
| "dataPath": "params_shard_27.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33431552, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.4.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.4.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 2621440 | |
| }, | |
| { | |
| "name": "language_model.model.layers.4.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 2949120 | |
| }, | |
| { | |
| "name": "language_model.model.layers.4.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 5046272 | |
| }, | |
| { | |
| "name": "language_model.model.layers.5.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 5308416 | |
| }, | |
| { | |
| "name": "language_model.model.layers.5.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 5312512 | |
| }, | |
| { | |
| "name": "language_model.model.layers.5.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 22089728 | |
| }, | |
| { | |
| "name": "language_model.model.layers.5.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 24186880 | |
| }, | |
| { | |
| "name": "language_model.model.layers.5.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 28381184 | |
| }, | |
| { | |
| "name": "language_model.model.layers.5.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 28385280 | |
| }, | |
| { | |
| "name": "language_model.model.layers.5.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 31006720 | |
| }, | |
| { | |
| "name": "language_model.model.layers.5.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 31334400 | |
| } | |
| ], | |
| "md5sum": "078c3506465d4ea7efc85d1f365ea4cc" | |
| }, | |
| { | |
| "dataPath": "params_shard_28.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.6.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "5952e83ee728812f4d91c95af0278f49" | |
| }, | |
| { | |
| "dataPath": "params_shard_29.bin", | |
| "format": "raw-shard", | |
| "nbytes": 16777216, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.7.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "4a0c0aee5bd65c3408e295d22864655c" | |
| }, | |
| { | |
| "dataPath": "params_shard_30.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.7.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "328e318b6751f17a691825e7a56c2eb7" | |
| }, | |
| { | |
| "dataPath": "params_shard_31.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30748672, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.5.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.6.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 262144 | |
| }, | |
| { | |
| "name": "language_model.model.layers.6.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 266240 | |
| }, | |
| { | |
| "name": "language_model.model.layers.6.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 17043456 | |
| }, | |
| { | |
| "name": "language_model.model.layers.6.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 19140608 | |
| }, | |
| { | |
| "name": "language_model.model.layers.6.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 23334912 | |
| }, | |
| { | |
| "name": "language_model.model.layers.6.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 23339008 | |
| }, | |
| { | |
| "name": "language_model.model.layers.6.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 25960448 | |
| }, | |
| { | |
| "name": "language_model.model.layers.6.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 26288128 | |
| }, | |
| { | |
| "name": "language_model.model.layers.6.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 28385280 | |
| }, | |
| { | |
| "name": "language_model.model.layers.7.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 28647424 | |
| }, | |
| { | |
| "name": "language_model.model.layers.7.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 28651520 | |
| } | |
| ], | |
| "md5sum": "e6f53b238749ed9fea6f468dfacf613a" | |
| }, | |
| { | |
| "dataPath": "params_shard_32.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.8.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "7b2561a8dcae3acac2ba199993c31ef7" | |
| }, | |
| { | |
| "dataPath": "params_shard_33.bin", | |
| "format": "raw-shard", | |
| "nbytes": 32583680, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.7.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.7.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 4194304 | |
| }, | |
| { | |
| "name": "language_model.model.layers.7.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 4198400 | |
| }, | |
| { | |
| "name": "language_model.model.layers.7.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 6819840 | |
| }, | |
| { | |
| "name": "language_model.model.layers.7.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 7147520 | |
| }, | |
| { | |
| "name": "language_model.model.layers.7.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 9244672 | |
| }, | |
| { | |
| "name": "language_model.model.layers.8.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 9506816 | |
| }, | |
| { | |
| "name": "language_model.model.layers.8.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 9510912 | |
| }, | |
| { | |
| "name": "language_model.model.layers.8.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 26288128 | |
| }, | |
| { | |
| "name": "language_model.model.layers.8.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 28385280 | |
| }, | |
| { | |
| "name": "language_model.model.layers.8.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 32579584 | |
| } | |
| ], | |
| "md5sum": "b9a048f98aa61a9f770184fe47bbb105" | |
| }, | |
| { | |
| "dataPath": "params_shard_34.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.9.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "891c1afba457c6fe64722ba549060f21" | |
| }, | |
| { | |
| "dataPath": "params_shard_35.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33431552, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.8.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "language_model.model.layers.8.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 2621440 | |
| }, | |
| { | |
| "name": "language_model.model.layers.8.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 2949120 | |
| }, | |
| { | |
| "name": "language_model.model.layers.8.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 5046272 | |
| }, | |
| { | |
| "name": "language_model.model.layers.9.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 5308416 | |
| }, | |
| { | |
| "name": "language_model.model.layers.9.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 5312512 | |
| }, | |
| { | |
| "name": "language_model.model.layers.9.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 22089728 | |
| }, | |
| { | |
| "name": "language_model.model.layers.9.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 24186880 | |
| }, | |
| { | |
| "name": "language_model.model.layers.9.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 28381184 | |
| }, | |
| { | |
| "name": "language_model.model.layers.9.self_attn.qkv_proj.q_weight", | |
| "shape": [ | |
| 2560, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2621440, | |
| "byteOffset": 28385280 | |
| }, | |
| { | |
| "name": "language_model.model.layers.9.self_attn.qkv_proj.q_scale", | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 327680, | |
| "byteOffset": 31006720 | |
| }, | |
| { | |
| "name": "language_model.model.layers.9.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 31334400 | |
| } | |
| ], | |
| "md5sum": "3a8c93c1595d3a0de84fc525432ff10c" | |
| }, | |
| { | |
| "dataPath": "params_shard_36.bin", | |
| "format": "raw-shard", | |
| "nbytes": 32801904, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.9.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "multi_modal_projector.linear_1.bias", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 262144 | |
| }, | |
| { | |
| "name": "multi_modal_projector.linear_1.q_weight", | |
| "shape": [ | |
| 2048, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 786432, | |
| "byteOffset": 266240 | |
| }, | |
| { | |
| "name": "multi_modal_projector.linear_1.q_scale", | |
| "shape": [ | |
| 2048, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 98304, | |
| "byteOffset": 1052672 | |
| }, | |
| { | |
| "name": "multi_modal_projector.linear_2.bias", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 1150976 | |
| }, | |
| { | |
| "name": "multi_modal_projector.linear_2.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 1155072 | |
| }, | |
| { | |
| "name": "multi_modal_projector.linear_2.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 3252224 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.embeddings.class_embedding", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 3514368 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.embeddings.patch_embedding.weight", | |
| "shape": [ | |
| 768, | |
| 3, | |
| 16, | |
| 16 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 3515904 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.embeddings.position_embedding.q_weight", | |
| "shape": [ | |
| 197, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 75648, | |
| "byteOffset": 4695552 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.embeddings.position_embedding.q_scale", | |
| "shape": [ | |
| 197, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 9456, | |
| "byteOffset": 4771200 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.layer_norm1.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 4780656 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.layer_norm1.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 4782192 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.layer_norm2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 4783728 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.layer_norm2.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 4785264 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.mlp.fc1.bias", | |
| "shape": [ | |
| 3072 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6144, | |
| "byteOffset": 4786800 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.mlp.fc1.q_weight", | |
| "shape": [ | |
| 3072, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 4792944 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.mlp.fc1.q_scale", | |
| "shape": [ | |
| 3072, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 5972592 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.mlp.fc2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 6120048 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.mlp.fc2.q_weight", | |
| "shape": [ | |
| 768, | |
| 384 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 6121584 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.mlp.fc2.q_scale", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 7301232 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.self_attn.k_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 7448688 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.self_attn.k_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 7450224 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.self_attn.k_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 7745136 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.self_attn.out_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 7782000 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.self_attn.out_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 7783536 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.self_attn.out_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 8078448 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.self_attn.q_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 8115312 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.self_attn.q_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 8116848 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.self_attn.q_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 8411760 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.self_attn.v_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 8448624 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.self_attn.v_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 8450160 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.0.self_attn.v_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 8745072 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.layer_norm1.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 8781936 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.layer_norm1.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 8783472 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.layer_norm2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 8785008 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.layer_norm2.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 8786544 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.mlp.fc1.bias", | |
| "shape": [ | |
| 3072 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6144, | |
| "byteOffset": 8788080 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.mlp.fc1.q_weight", | |
| "shape": [ | |
| 3072, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 8794224 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.mlp.fc1.q_scale", | |
| "shape": [ | |
| 3072, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 9973872 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.mlp.fc2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 10121328 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.mlp.fc2.q_weight", | |
| "shape": [ | |
| 768, | |
| 384 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 10122864 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.mlp.fc2.q_scale", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 11302512 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.self_attn.k_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 11449968 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.self_attn.k_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 11451504 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.self_attn.k_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 11746416 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.self_attn.out_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 11783280 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.self_attn.out_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 11784816 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.self_attn.out_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 12079728 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.self_attn.q_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 12116592 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.self_attn.q_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 12118128 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.self_attn.q_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 12413040 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.self_attn.v_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 12449904 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.self_attn.v_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 12451440 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.1.self_attn.v_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 12746352 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.layer_norm1.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 12783216 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.layer_norm1.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 12784752 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.layer_norm2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 12786288 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.layer_norm2.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 12787824 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.mlp.fc1.bias", | |
| "shape": [ | |
| 3072 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6144, | |
| "byteOffset": 12789360 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.mlp.fc1.q_weight", | |
| "shape": [ | |
| 3072, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 12795504 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.mlp.fc1.q_scale", | |
| "shape": [ | |
| 3072, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 13975152 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.mlp.fc2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 14122608 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.mlp.fc2.q_weight", | |
| "shape": [ | |
| 768, | |
| 384 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 14124144 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.mlp.fc2.q_scale", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 15303792 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.self_attn.k_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 15451248 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.self_attn.k_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 15452784 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.self_attn.k_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 15747696 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.self_attn.out_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 15784560 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.self_attn.out_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 15786096 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.self_attn.out_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 16081008 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.self_attn.q_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 16117872 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.self_attn.q_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 16119408 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.self_attn.q_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 16414320 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.self_attn.v_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 16451184 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.self_attn.v_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 16452720 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.10.self_attn.v_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 16747632 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.layer_norm1.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 16784496 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.layer_norm1.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 16786032 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.layer_norm2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 16787568 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.layer_norm2.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 16789104 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.mlp.fc1.bias", | |
| "shape": [ | |
| 3072 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6144, | |
| "byteOffset": 16790640 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.mlp.fc1.q_weight", | |
| "shape": [ | |
| 3072, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 16796784 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.mlp.fc1.q_scale", | |
| "shape": [ | |
| 3072, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 17976432 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.mlp.fc2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 18123888 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.mlp.fc2.q_weight", | |
| "shape": [ | |
| 768, | |
| 384 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 18125424 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.mlp.fc2.q_scale", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 19305072 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.self_attn.k_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 19452528 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.self_attn.k_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 19454064 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.self_attn.k_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 19748976 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.self_attn.out_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 19785840 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.self_attn.out_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 19787376 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.self_attn.out_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 20082288 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.self_attn.q_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 20119152 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.self_attn.q_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 20120688 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.self_attn.q_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 20415600 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.self_attn.v_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 20452464 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.self_attn.v_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 20454000 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.11.self_attn.v_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 20748912 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.layer_norm1.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 20785776 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.layer_norm1.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 20787312 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.layer_norm2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 20788848 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.layer_norm2.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 20790384 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.mlp.fc1.bias", | |
| "shape": [ | |
| 3072 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6144, | |
| "byteOffset": 20791920 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.mlp.fc1.q_weight", | |
| "shape": [ | |
| 3072, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 20798064 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.mlp.fc1.q_scale", | |
| "shape": [ | |
| 3072, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 21977712 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.mlp.fc2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 22125168 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.mlp.fc2.q_weight", | |
| "shape": [ | |
| 768, | |
| 384 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 22126704 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.mlp.fc2.q_scale", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 23306352 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.self_attn.k_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 23453808 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.self_attn.k_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 23455344 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.self_attn.k_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 23750256 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.self_attn.out_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 23787120 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.self_attn.out_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 23788656 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.self_attn.out_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 24083568 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.self_attn.q_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 24120432 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.self_attn.q_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 24121968 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.self_attn.q_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 24416880 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.self_attn.v_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 24453744 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.self_attn.v_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 24455280 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.2.self_attn.v_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 24750192 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.layer_norm1.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 24787056 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.layer_norm1.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 24788592 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.layer_norm2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 24790128 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.layer_norm2.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 24791664 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.mlp.fc1.bias", | |
| "shape": [ | |
| 3072 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6144, | |
| "byteOffset": 24793200 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.mlp.fc1.q_weight", | |
| "shape": [ | |
| 3072, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 24799344 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.mlp.fc1.q_scale", | |
| "shape": [ | |
| 3072, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 25978992 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.mlp.fc2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 26126448 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.mlp.fc2.q_weight", | |
| "shape": [ | |
| 768, | |
| 384 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 26127984 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.mlp.fc2.q_scale", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 27307632 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.self_attn.k_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 27455088 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.self_attn.k_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 27456624 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.self_attn.k_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 27751536 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.self_attn.out_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 27788400 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.self_attn.out_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 27789936 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.self_attn.out_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 28084848 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.self_attn.q_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 28121712 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.self_attn.q_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 28123248 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.self_attn.q_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 28418160 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.self_attn.v_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 28455024 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.self_attn.v_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 28456560 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.3.self_attn.v_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 28751472 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.layer_norm1.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 28788336 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.layer_norm1.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 28789872 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.layer_norm2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 28791408 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.layer_norm2.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 28792944 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.mlp.fc1.bias", | |
| "shape": [ | |
| 3072 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6144, | |
| "byteOffset": 28794480 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.mlp.fc1.q_weight", | |
| "shape": [ | |
| 3072, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 28800624 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.mlp.fc1.q_scale", | |
| "shape": [ | |
| 3072, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 29980272 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.mlp.fc2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 30127728 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.mlp.fc2.q_weight", | |
| "shape": [ | |
| 768, | |
| 384 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 30129264 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.mlp.fc2.q_scale", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 31308912 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.self_attn.k_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 31456368 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.self_attn.k_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 31457904 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.self_attn.k_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 31752816 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.self_attn.out_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 31789680 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.self_attn.out_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 31791216 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.self_attn.out_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 32086128 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.self_attn.q_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 32122992 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.self_attn.q_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 32124528 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.self_attn.q_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 32419440 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.self_attn.v_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 32456304 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.self_attn.v_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 32457840 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.4.self_attn.v_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 32752752 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.layer_norm1.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 32789616 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.layer_norm1.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 32791152 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.layer_norm2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 32792688 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.layer_norm2.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 32794224 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.mlp.fc1.bias", | |
| "shape": [ | |
| 3072 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6144, | |
| "byteOffset": 32795760 | |
| } | |
| ], | |
| "md5sum": "da721a7f6fdb0bb1aa4e2d0ed99d8949" | |
| }, | |
| { | |
| "dataPath": "params_shard_37.bin", | |
| "format": "raw-shard", | |
| "nbytes": 16777216, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.17.mlp.down_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 16777216, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "23e9b191c363c33250d1f39122ae7380" | |
| }, | |
| { | |
| "dataPath": "params_shard_38.bin", | |
| "format": "raw-shard", | |
| "nbytes": 33554432, | |
| "records": [ | |
| { | |
| "name": "language_model.model.layers.17.mlp.gate_up_proj.q_weight", | |
| "shape": [ | |
| 32768, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 33554432, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "de1b5f5a3309cb832f70c9dabecd6659" | |
| }, | |
| { | |
| "dataPath": "params_shard_39.bin", | |
| "format": "raw-shard", | |
| "nbytes": 28663296, | |
| "records": [ | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.mlp.fc1.q_weight", | |
| "shape": [ | |
| 3072, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.mlp.fc1.q_scale", | |
| "shape": [ | |
| 3072, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 1179648 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.mlp.fc2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 1327104 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.mlp.fc2.q_weight", | |
| "shape": [ | |
| 768, | |
| 384 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 1328640 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.mlp.fc2.q_scale", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 2508288 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.self_attn.k_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 2655744 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.self_attn.k_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 2657280 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.self_attn.k_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 2952192 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.self_attn.out_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 2989056 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.self_attn.out_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 2990592 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.self_attn.out_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 3285504 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.self_attn.q_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 3322368 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.self_attn.q_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 3323904 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.self_attn.q_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 3618816 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.self_attn.v_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 3655680 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.self_attn.v_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 3657216 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.5.self_attn.v_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 3952128 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.layer_norm1.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 3988992 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.layer_norm1.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 3990528 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.layer_norm2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 3992064 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.layer_norm2.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 3993600 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.mlp.fc1.bias", | |
| "shape": [ | |
| 3072 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6144, | |
| "byteOffset": 3995136 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.mlp.fc1.q_weight", | |
| "shape": [ | |
| 3072, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 4001280 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.mlp.fc1.q_scale", | |
| "shape": [ | |
| 3072, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 5180928 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.mlp.fc2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 5328384 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.mlp.fc2.q_weight", | |
| "shape": [ | |
| 768, | |
| 384 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 5329920 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.mlp.fc2.q_scale", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 6509568 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.self_attn.k_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 6657024 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.self_attn.k_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 6658560 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.self_attn.k_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 6953472 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.self_attn.out_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 6990336 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.self_attn.out_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 6991872 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.self_attn.out_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 7286784 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.self_attn.q_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 7323648 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.self_attn.q_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 7325184 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.self_attn.q_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 7620096 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.self_attn.v_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 7656960 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.self_attn.v_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 7658496 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.6.self_attn.v_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 7953408 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.layer_norm1.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 7990272 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.layer_norm1.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 7991808 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.layer_norm2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 7993344 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.layer_norm2.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 7994880 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.mlp.fc1.bias", | |
| "shape": [ | |
| 3072 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6144, | |
| "byteOffset": 7996416 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.mlp.fc1.q_weight", | |
| "shape": [ | |
| 3072, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 8002560 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.mlp.fc1.q_scale", | |
| "shape": [ | |
| 3072, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 9182208 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.mlp.fc2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 9329664 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.mlp.fc2.q_weight", | |
| "shape": [ | |
| 768, | |
| 384 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 9331200 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.mlp.fc2.q_scale", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 10510848 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.self_attn.k_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 10658304 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.self_attn.k_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 10659840 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.self_attn.k_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 10954752 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.self_attn.out_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 10991616 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.self_attn.out_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 10993152 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.self_attn.out_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 11288064 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.self_attn.q_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 11324928 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.self_attn.q_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 11326464 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.self_attn.q_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 11621376 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.self_attn.v_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 11658240 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.self_attn.v_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 11659776 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.7.self_attn.v_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 11954688 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.layer_norm1.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 11991552 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.layer_norm1.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 11993088 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.layer_norm2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 11994624 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.layer_norm2.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 11996160 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.mlp.fc1.bias", | |
| "shape": [ | |
| 3072 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6144, | |
| "byteOffset": 11997696 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.mlp.fc1.q_weight", | |
| "shape": [ | |
| 3072, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 12003840 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.mlp.fc1.q_scale", | |
| "shape": [ | |
| 3072, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 13183488 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.mlp.fc2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 13330944 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.mlp.fc2.q_weight", | |
| "shape": [ | |
| 768, | |
| 384 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 13332480 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.mlp.fc2.q_scale", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 14512128 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.self_attn.k_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 14659584 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.self_attn.k_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 14661120 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.self_attn.k_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 14956032 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.self_attn.out_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 14992896 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.self_attn.out_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 14994432 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.self_attn.out_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 15289344 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.self_attn.q_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 15326208 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.self_attn.q_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 15327744 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.self_attn.q_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 15622656 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.self_attn.v_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 15659520 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.self_attn.v_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 15661056 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.8.self_attn.v_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 15955968 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.layer_norm1.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 15992832 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.layer_norm1.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 15994368 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.layer_norm2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 15995904 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.layer_norm2.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 15997440 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.mlp.fc1.bias", | |
| "shape": [ | |
| 3072 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 6144, | |
| "byteOffset": 15998976 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.mlp.fc1.q_weight", | |
| "shape": [ | |
| 3072, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 16005120 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.mlp.fc1.q_scale", | |
| "shape": [ | |
| 3072, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 17184768 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.mlp.fc2.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 17332224 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.mlp.fc2.q_weight", | |
| "shape": [ | |
| 768, | |
| 384 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1179648, | |
| "byteOffset": 17333760 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.mlp.fc2.q_scale", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 147456, | |
| "byteOffset": 18513408 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.self_attn.k_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 18660864 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.self_attn.k_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 18662400 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.self_attn.k_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 18957312 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.self_attn.out_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 18994176 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.self_attn.out_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 18995712 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.self_attn.out_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 19290624 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.self_attn.q_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 19327488 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.self_attn.q_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 19329024 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.self_attn.q_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 19623936 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.self_attn.v_proj.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 19660800 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.self_attn.v_proj.q_weight", | |
| "shape": [ | |
| 768, | |
| 96 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 294912, | |
| "byteOffset": 19662336 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.encoder.layers.9.self_attn.v_proj.q_scale", | |
| "shape": [ | |
| 768, | |
| 24 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 36864, | |
| "byteOffset": 19957248 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.post_layernorm.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 19994112 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.post_layernorm.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 19995648 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.pre_layrnorm.bias", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 19997184 | |
| }, | |
| { | |
| "name": "vision_tower.vision_model.pre_layrnorm.weight", | |
| "shape": [ | |
| 768 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 1536, | |
| "byteOffset": 19998720 | |
| }, | |
| { | |
| "name": "language_model.model.layers.17.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 20000256 | |
| }, | |
| { | |
| "name": "language_model.model.layers.17.mlp.down_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 512 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 20004352 | |
| }, | |
| { | |
| "name": "language_model.model.layers.17.mlp.gate_up_proj.q_scale", | |
| "shape": [ | |
| 32768, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4194304, | |
| "byteOffset": 22101504 | |
| }, | |
| { | |
| "name": "language_model.model.layers.17.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 26295808 | |
| }, | |
| { | |
| "name": "language_model.model.layers.17.self_attn.o_proj.q_weight", | |
| "shape": [ | |
| 2048, | |
| 256 | |
| ], | |
| "dtype": "uint32", | |
| "format": "f32-to-bf16", | |
| "nbytes": 2097152, | |
| "byteOffset": 26299904 | |
| }, | |
| { | |
| "name": "language_model.model.layers.17.self_attn.o_proj.q_scale", | |
| "shape": [ | |
| 2048, | |
| 64 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 262144, | |
| "byteOffset": 28397056 | |
| }, | |
| { | |
| "name": "language_model.model.norm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 28659200 | |
| } | |
| ], | |
| "md5sum": "68ffec0fce25fd72b5892e3b23fb2b99" | |
| } | |
| ] | |
| } |