| { |
| "metadata": { |
| "ParamSize": 165, |
| "ParamBytes": 319491072.0, |
| "BitsPerParam": 4.5010265163093015 |
| }, |
| "records": [ |
| { |
| "dataPath": "params_shard_0.bin", |
| "format": "raw-shard", |
| "nbytes": 32962560, |
| "records": [ |
| { |
| "name": "model.embed_tokens.q_weight", |
| "shape": [ |
| 32000, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 24576000, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.embed_tokens.q_scale", |
| "shape": [ |
| 32000, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072000, |
| "byteOffset": 24576000 |
| }, |
| { |
| "name": "model.layers.0.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 27648000 |
| }, |
| { |
| "name": "model.layers.0.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 27651072 |
| }, |
| { |
| "name": "model.layers.0.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 31190016 |
| }, |
| { |
| "name": "model.layers.0.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 31632384 |
| }, |
| { |
| "name": "model.layers.0.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 32812032 |
| }, |
| { |
| "name": "model.layers.0.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 32959488 |
| } |
| ], |
| "md5sum": "9ca3196d92936403df3a409c94160c7e" |
| }, |
| { |
| "dataPath": "params_shard_1.bin", |
| "format": "raw-shard", |
| "nbytes": 33024000, |
| "records": [ |
| { |
| "name": "model.layers.0.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.0.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 6635520 |
| }, |
| { |
| "name": "model.layers.0.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 7464960 |
| }, |
| { |
| "name": "model.layers.0.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 10782720 |
| }, |
| { |
| "name": "model.layers.1.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 11197440 |
| }, |
| { |
| "name": "model.layers.1.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 11200512 |
| }, |
| { |
| "name": "model.layers.1.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 14739456 |
| }, |
| { |
| "name": "model.layers.1.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 15181824 |
| }, |
| { |
| "name": "model.layers.1.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 16361472 |
| }, |
| { |
| "name": "model.layers.1.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 16508928 |
| }, |
| { |
| "name": "model.layers.1.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 16512000 |
| }, |
| { |
| "name": "model.layers.1.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 23147520 |
| }, |
| { |
| "name": "model.layers.1.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 23976960 |
| }, |
| { |
| "name": "model.layers.1.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 27294720 |
| }, |
| { |
| "name": "model.layers.2.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 27709440 |
| }, |
| { |
| "name": "model.layers.2.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 27712512 |
| }, |
| { |
| "name": "model.layers.2.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 31251456 |
| }, |
| { |
| "name": "model.layers.2.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 31693824 |
| }, |
| { |
| "name": "model.layers.2.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 32873472 |
| }, |
| { |
| "name": "model.layers.2.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 33020928 |
| } |
| ], |
| "md5sum": "7bdb8681330907ec9061400cb44f7d8e" |
| }, |
| { |
| "dataPath": "params_shard_2.bin", |
| "format": "raw-shard", |
| "nbytes": 33024000, |
| "records": [ |
| { |
| "name": "model.layers.2.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.2.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 6635520 |
| }, |
| { |
| "name": "model.layers.2.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 7464960 |
| }, |
| { |
| "name": "model.layers.2.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 10782720 |
| }, |
| { |
| "name": "model.layers.3.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 11197440 |
| }, |
| { |
| "name": "model.layers.3.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 11200512 |
| }, |
| { |
| "name": "model.layers.3.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 14739456 |
| }, |
| { |
| "name": "model.layers.3.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 15181824 |
| }, |
| { |
| "name": "model.layers.3.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 16361472 |
| }, |
| { |
| "name": "model.layers.3.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 16508928 |
| }, |
| { |
| "name": "model.layers.3.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 16512000 |
| }, |
| { |
| "name": "model.layers.3.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 23147520 |
| }, |
| { |
| "name": "model.layers.3.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 23976960 |
| }, |
| { |
| "name": "model.layers.3.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 27294720 |
| }, |
| { |
| "name": "model.layers.4.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 27709440 |
| }, |
| { |
| "name": "model.layers.4.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 27712512 |
| }, |
| { |
| "name": "model.layers.4.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 31251456 |
| }, |
| { |
| "name": "model.layers.4.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 31693824 |
| }, |
| { |
| "name": "model.layers.4.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 32873472 |
| }, |
| { |
| "name": "model.layers.4.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 33020928 |
| } |
| ], |
| "md5sum": "d3e47c0eccbf600e95e382b574e94238" |
| }, |
| { |
| "dataPath": "params_shard_3.bin", |
| "format": "raw-shard", |
| "nbytes": 33024000, |
| "records": [ |
| { |
| "name": "model.layers.4.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.4.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 6635520 |
| }, |
| { |
| "name": "model.layers.4.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 7464960 |
| }, |
| { |
| "name": "model.layers.4.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 10782720 |
| }, |
| { |
| "name": "model.layers.5.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 11197440 |
| }, |
| { |
| "name": "model.layers.5.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 11200512 |
| }, |
| { |
| "name": "model.layers.5.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 14739456 |
| }, |
| { |
| "name": "model.layers.5.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 15181824 |
| }, |
| { |
| "name": "model.layers.5.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 16361472 |
| }, |
| { |
| "name": "model.layers.5.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 16508928 |
| }, |
| { |
| "name": "model.layers.5.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 16512000 |
| }, |
| { |
| "name": "model.layers.5.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 23147520 |
| }, |
| { |
| "name": "model.layers.5.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 23976960 |
| }, |
| { |
| "name": "model.layers.5.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 27294720 |
| }, |
| { |
| "name": "model.layers.6.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 27709440 |
| }, |
| { |
| "name": "model.layers.6.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 27712512 |
| }, |
| { |
| "name": "model.layers.6.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 31251456 |
| }, |
| { |
| "name": "model.layers.6.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 31693824 |
| }, |
| { |
| "name": "model.layers.6.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 32873472 |
| }, |
| { |
| "name": "model.layers.6.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 33020928 |
| } |
| ], |
| "md5sum": "ab1e2b7651ed778cc99544da6fd3bcfa" |
| }, |
| { |
| "dataPath": "params_shard_4.bin", |
| "format": "raw-shard", |
| "nbytes": 33024000, |
| "records": [ |
| { |
| "name": "model.layers.6.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.6.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 6635520 |
| }, |
| { |
| "name": "model.layers.6.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 7464960 |
| }, |
| { |
| "name": "model.layers.6.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 10782720 |
| }, |
| { |
| "name": "model.layers.7.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 11197440 |
| }, |
| { |
| "name": "model.layers.7.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 11200512 |
| }, |
| { |
| "name": "model.layers.7.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 14739456 |
| }, |
| { |
| "name": "model.layers.7.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 15181824 |
| }, |
| { |
| "name": "model.layers.7.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 16361472 |
| }, |
| { |
| "name": "model.layers.7.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 16508928 |
| }, |
| { |
| "name": "model.layers.7.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 16512000 |
| }, |
| { |
| "name": "model.layers.7.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 23147520 |
| }, |
| { |
| "name": "model.layers.7.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 23976960 |
| }, |
| { |
| "name": "model.layers.7.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 27294720 |
| }, |
| { |
| "name": "model.layers.8.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 27709440 |
| }, |
| { |
| "name": "model.layers.8.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 27712512 |
| }, |
| { |
| "name": "model.layers.8.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 31251456 |
| }, |
| { |
| "name": "model.layers.8.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 31693824 |
| }, |
| { |
| "name": "model.layers.8.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 32873472 |
| }, |
| { |
| "name": "model.layers.8.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 33020928 |
| } |
| ], |
| "md5sum": "d4b6c15d1bb4711f6e9b8c8286e4130c" |
| }, |
| { |
| "dataPath": "params_shard_5.bin", |
| "format": "raw-shard", |
| "nbytes": 33024000, |
| "records": [ |
| { |
| "name": "model.layers.8.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.8.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 6635520 |
| }, |
| { |
| "name": "model.layers.8.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 7464960 |
| }, |
| { |
| "name": "model.layers.8.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 10782720 |
| }, |
| { |
| "name": "model.layers.9.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 11197440 |
| }, |
| { |
| "name": "model.layers.9.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 11200512 |
| }, |
| { |
| "name": "model.layers.9.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 14739456 |
| }, |
| { |
| "name": "model.layers.9.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 15181824 |
| }, |
| { |
| "name": "model.layers.9.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 16361472 |
| }, |
| { |
| "name": "model.layers.9.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 16508928 |
| }, |
| { |
| "name": "model.layers.9.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 16512000 |
| }, |
| { |
| "name": "model.layers.9.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 23147520 |
| }, |
| { |
| "name": "model.layers.9.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 23976960 |
| }, |
| { |
| "name": "model.layers.9.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 27294720 |
| }, |
| { |
| "name": "model.layers.10.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 27709440 |
| }, |
| { |
| "name": "model.layers.10.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 27712512 |
| }, |
| { |
| "name": "model.layers.10.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 31251456 |
| }, |
| { |
| "name": "model.layers.10.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 31693824 |
| }, |
| { |
| "name": "model.layers.10.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 32873472 |
| }, |
| { |
| "name": "model.layers.10.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 33020928 |
| } |
| ], |
| "md5sum": "eac494812b1c9dfd2f6dde28b205cc6c" |
| }, |
| { |
| "dataPath": "params_shard_6.bin", |
| "format": "raw-shard", |
| "nbytes": 33024000, |
| "records": [ |
| { |
| "name": "model.layers.10.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.10.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 6635520 |
| }, |
| { |
| "name": "model.layers.10.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 7464960 |
| }, |
| { |
| "name": "model.layers.10.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 10782720 |
| }, |
| { |
| "name": "model.layers.11.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 11197440 |
| }, |
| { |
| "name": "model.layers.11.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 11200512 |
| }, |
| { |
| "name": "model.layers.11.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 14739456 |
| }, |
| { |
| "name": "model.layers.11.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 15181824 |
| }, |
| { |
| "name": "model.layers.11.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 16361472 |
| }, |
| { |
| "name": "model.layers.11.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 16508928 |
| }, |
| { |
| "name": "model.layers.11.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 16512000 |
| }, |
| { |
| "name": "model.layers.11.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 23147520 |
| }, |
| { |
| "name": "model.layers.11.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 23976960 |
| }, |
| { |
| "name": "model.layers.11.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 27294720 |
| }, |
| { |
| "name": "model.layers.12.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 27709440 |
| }, |
| { |
| "name": "model.layers.12.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 27712512 |
| }, |
| { |
| "name": "model.layers.12.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 31251456 |
| }, |
| { |
| "name": "model.layers.12.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 31693824 |
| }, |
| { |
| "name": "model.layers.12.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 32873472 |
| }, |
| { |
| "name": "model.layers.12.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 33020928 |
| } |
| ], |
| "md5sum": "486523010c2b81edb76e8ff9f5d58916" |
| }, |
| { |
| "dataPath": "params_shard_7.bin", |
| "format": "raw-shard", |
| "nbytes": 33024000, |
| "records": [ |
| { |
| "name": "model.layers.12.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.12.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 6635520 |
| }, |
| { |
| "name": "model.layers.12.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 7464960 |
| }, |
| { |
| "name": "model.layers.12.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 10782720 |
| }, |
| { |
| "name": "model.layers.13.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 11197440 |
| }, |
| { |
| "name": "model.layers.13.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 11200512 |
| }, |
| { |
| "name": "model.layers.13.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 14739456 |
| }, |
| { |
| "name": "model.layers.13.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 15181824 |
| }, |
| { |
| "name": "model.layers.13.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 16361472 |
| }, |
| { |
| "name": "model.layers.13.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 16508928 |
| }, |
| { |
| "name": "model.layers.13.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 16512000 |
| }, |
| { |
| "name": "model.layers.13.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 23147520 |
| }, |
| { |
| "name": "model.layers.13.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 23976960 |
| }, |
| { |
| "name": "model.layers.13.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 27294720 |
| }, |
| { |
| "name": "model.layers.14.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 27709440 |
| }, |
| { |
| "name": "model.layers.14.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 27712512 |
| }, |
| { |
| "name": "model.layers.14.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 31251456 |
| }, |
| { |
| "name": "model.layers.14.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 31693824 |
| }, |
| { |
| "name": "model.layers.14.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 32873472 |
| }, |
| { |
| "name": "model.layers.14.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 33020928 |
| } |
| ], |
| "md5sum": "8d03ff008c9bed110c2eef169d1d312f" |
| }, |
| { |
| "dataPath": "params_shard_8.bin", |
| "format": "raw-shard", |
| "nbytes": 24576000, |
| "records": [ |
| { |
| "name": "lm_head.q_weight", |
| "shape": [ |
| 32000, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 24576000, |
| "byteOffset": 0 |
| } |
| ], |
| "md5sum": "90eca5b898141bb2381906eb86e353df" |
| }, |
| { |
| "dataPath": "params_shard_9.bin", |
| "format": "raw-shard", |
| "nbytes": 30784512, |
| "records": [ |
| { |
| "name": "model.layers.14.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 0 |
| }, |
| { |
| "name": "model.layers.14.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 6635520 |
| }, |
| { |
| "name": "model.layers.14.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 7464960 |
| }, |
| { |
| "name": "model.layers.14.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 10782720 |
| }, |
| { |
| "name": "model.layers.15.input_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 11197440 |
| }, |
| { |
| "name": "model.layers.15.self_attn.qkv_proj.q_weight", |
| "shape": [ |
| 4608, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3538944, |
| "byteOffset": 11200512 |
| }, |
| { |
| "name": "model.layers.15.self_attn.qkv_proj.q_scale", |
| "shape": [ |
| 4608, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 442368, |
| "byteOffset": 14739456 |
| }, |
| { |
| "name": "model.layers.15.self_attn.o_proj.q_weight", |
| "shape": [ |
| 1536, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 1179648, |
| "byteOffset": 15181824 |
| }, |
| { |
| "name": "model.layers.15.self_attn.o_proj.q_scale", |
| "shape": [ |
| 1536, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 147456, |
| "byteOffset": 16361472 |
| }, |
| { |
| "name": "model.layers.15.post_attention_layernorm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 16508928 |
| }, |
| { |
| "name": "model.layers.15.mlp.gate_up_proj.q_weight", |
| "shape": [ |
| 8640, |
| 192 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 6635520, |
| "byteOffset": 16512000 |
| }, |
| { |
| "name": "model.layers.15.mlp.gate_up_proj.q_scale", |
| "shape": [ |
| 8640, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 829440, |
| "byteOffset": 23147520 |
| }, |
| { |
| "name": "model.layers.15.mlp.down_proj.q_weight", |
| "shape": [ |
| 1536, |
| 540 |
| ], |
| "dtype": "uint32", |
| "format": "f32-to-bf16", |
| "nbytes": 3317760, |
| "byteOffset": 23976960 |
| }, |
| { |
| "name": "model.layers.15.mlp.down_proj.q_scale", |
| "shape": [ |
| 1536, |
| 135 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 414720, |
| "byteOffset": 27294720 |
| }, |
| { |
| "name": "model.norm.weight", |
| "shape": [ |
| 1536 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072, |
| "byteOffset": 27709440 |
| }, |
| { |
| "name": "lm_head.q_scale", |
| "shape": [ |
| 32000, |
| 48 |
| ], |
| "dtype": "float16", |
| "format": "f32-to-bf16", |
| "nbytes": 3072000, |
| "byteOffset": 27712512 |
| } |
| ], |
| "md5sum": "cc25361e08b6e08fd245482aad0afb4d" |
| } |
| ] |
| } |