{ "metadata": { "ParamSize": 403, "ParamBytes": 1789382016.0, "BitsPerParam": 6.827912083021164 }, "records": [ { "dataPath": "params_shard_0.bin", "format": "raw-shard", "nbytes": 424258560, "records": [ { "name": "model.embed_tokens.weight", "shape": [ 122760, 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 424258560, "byteOffset": 0 } ], "md5sum": "fa76ad8440a74eb26c2d0ea5da598383" }, { "dataPath": "params_shard_1.bin", "format": "raw-shard", "nbytes": 30240000, "records": [ { "name": "model.layers.0.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 0 }, { "name": "model.layers.0.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 4478976 }, { "name": "model.layers.0.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 5038848 }, { "name": "model.layers.0.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 6531840 }, { "name": "model.layers.0.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 6718464 }, { "name": "model.layers.0.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 16671744 }, { "name": "model.layers.0.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 17915904 }, { "name": "model.layers.0.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 22892544 }, { "name": "model.layers.0.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23514624 }, { "name": "model.layers.0.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23518080 }, { "name": "model.layers.1.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 23521536 }, { "name": "model.layers.1.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 28000512 }, { "name": "model.layers.1.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 28560384 }, { "name": "model.layers.1.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 30053376 } ], "md5sum": "cf95ca2c59db18d7a26bfb930fdf1fdf" }, { "dataPath": "params_shard_2.bin", "format": "raw-shard", "nbytes": 33474816, "records": [ { "name": "model.layers.1.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 0 }, { "name": "model.layers.1.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 9953280 }, { "name": "model.layers.1.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 11197440 }, { "name": "model.layers.1.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 16174080 }, { "name": "model.layers.1.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16796160 }, { "name": "model.layers.1.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16799616 }, { "name": "model.layers.2.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 16803072 }, { "name": "model.layers.2.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 21282048 }, { "name": "model.layers.2.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 21841920 }, { "name": "model.layers.2.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 23334912 }, { "name": "model.layers.2.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 23521536 } ], "md5sum": "75847967acbe131c529a29f63d16ea48" }, { "dataPath": "params_shard_3.bin", "format": "raw-shard", "nbytes": 30371328, "records": [ { "name": "model.layers.2.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 0 }, { "name": "model.layers.2.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 1244160 }, { "name": "model.layers.2.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 6220800 }, { "name": "model.layers.2.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6842880 }, { "name": "model.layers.2.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6846336 }, { "name": "model.layers.3.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 6849792 }, { "name": "model.layers.3.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 11328768 }, { "name": "model.layers.3.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 11888640 }, { "name": "model.layers.3.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 13381632 }, { "name": "model.layers.3.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 13568256 }, { "name": "model.layers.3.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 23521536 }, { "name": "model.layers.3.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 24765696 }, { "name": "model.layers.3.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 29742336 }, { "name": "model.layers.3.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30364416 }, { "name": "model.layers.3.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30367872 } ], "md5sum": "bb633babd720fad47b2076128d72efbc" }, { "dataPath": "params_shard_4.bin", "format": "raw-shard", "nbytes": 30240000, "records": [ { "name": "model.layers.4.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 0 }, { "name": "model.layers.4.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 4478976 }, { "name": "model.layers.4.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 5038848 }, { "name": "model.layers.4.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 6531840 }, { "name": "model.layers.4.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 6718464 }, { "name": "model.layers.4.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 16671744 }, { "name": "model.layers.4.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 17915904 }, { "name": "model.layers.4.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 22892544 }, { "name": "model.layers.4.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23514624 }, { "name": "model.layers.4.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23518080 }, { "name": "model.layers.5.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 23521536 }, { "name": "model.layers.5.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 28000512 }, { "name": "model.layers.5.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 28560384 }, { "name": "model.layers.5.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 30053376 } ], "md5sum": "cd4cd49ca554579cf9826f5bddb1afd6" }, { "dataPath": "params_shard_5.bin", "format": "raw-shard", "nbytes": 33474816, "records": [ { "name": "model.layers.5.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 0 }, { "name": "model.layers.5.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 9953280 }, { "name": "model.layers.5.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 11197440 }, { "name": "model.layers.5.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 16174080 }, { "name": "model.layers.5.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16796160 }, { "name": "model.layers.5.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16799616 }, { "name": "model.layers.6.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 16803072 }, { "name": "model.layers.6.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 21282048 }, { "name": "model.layers.6.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 21841920 }, { "name": "model.layers.6.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 23334912 }, { "name": "model.layers.6.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 23521536 } ], "md5sum": "166f5dfb13ec0bdf83168672d8f8f1be" }, { "dataPath": "params_shard_6.bin", "format": "raw-shard", "nbytes": 30371328, "records": [ { "name": "model.layers.6.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 0 }, { "name": "model.layers.6.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 1244160 }, { "name": "model.layers.6.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 6220800 }, { "name": "model.layers.6.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6842880 }, { "name": "model.layers.6.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6846336 }, { "name": "model.layers.7.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 6849792 }, { "name": "model.layers.7.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 11328768 }, { "name": "model.layers.7.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 11888640 }, { "name": "model.layers.7.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 13381632 }, { "name": "model.layers.7.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 13568256 }, { "name": "model.layers.7.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 23521536 }, { "name": "model.layers.7.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 24765696 }, { "name": "model.layers.7.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 29742336 }, { "name": "model.layers.7.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30364416 }, { "name": "model.layers.7.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30367872 } ], "md5sum": "7eb43f0256206cd0905343d7ddecd0be" }, { "dataPath": "params_shard_7.bin", "format": "raw-shard", "nbytes": 30240000, "records": [ { "name": "model.layers.8.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 0 }, { "name": "model.layers.8.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 4478976 }, { "name": "model.layers.8.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 5038848 }, { "name": "model.layers.8.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 6531840 }, { "name": "model.layers.8.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 6718464 }, { "name": "model.layers.8.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 16671744 }, { "name": "model.layers.8.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 17915904 }, { "name": "model.layers.8.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 22892544 }, { "name": "model.layers.8.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23514624 }, { "name": "model.layers.8.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23518080 }, { "name": "model.layers.9.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 23521536 }, { "name": "model.layers.9.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 28000512 }, { "name": "model.layers.9.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 28560384 }, { "name": "model.layers.9.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 30053376 } ], "md5sum": "9cd28d6fc84c5a03c0a59e376c994bc7" }, { "dataPath": "params_shard_8.bin", "format": "raw-shard", "nbytes": 33474816, "records": [ { "name": "model.layers.9.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 0 }, { "name": "model.layers.9.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 9953280 }, { "name": "model.layers.9.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 11197440 }, { "name": "model.layers.9.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 16174080 }, { "name": "model.layers.9.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16796160 }, { "name": "model.layers.9.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16799616 }, { "name": "model.layers.10.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 16803072 }, { "name": "model.layers.10.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 21282048 }, { "name": "model.layers.10.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 21841920 }, { "name": "model.layers.10.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 23334912 }, { "name": "model.layers.10.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 23521536 } ], "md5sum": "b79dac393504983b352a38c7dd5988d9" }, { "dataPath": "params_shard_9.bin", "format": "raw-shard", "nbytes": 30371328, "records": [ { "name": "model.layers.10.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 0 }, { "name": "model.layers.10.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 1244160 }, { "name": "model.layers.10.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 6220800 }, { "name": "model.layers.10.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6842880 }, { "name": "model.layers.10.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6846336 }, { "name": "model.layers.11.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 6849792 }, { "name": "model.layers.11.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 11328768 }, { "name": "model.layers.11.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 11888640 }, { "name": "model.layers.11.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 13381632 }, { "name": "model.layers.11.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 13568256 }, { "name": "model.layers.11.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 23521536 }, { "name": "model.layers.11.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 24765696 }, { "name": "model.layers.11.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 29742336 }, { "name": "model.layers.11.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30364416 }, { "name": "model.layers.11.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30367872 } ], "md5sum": "733b9ade2e29dcc314762b7061d8bc83" }, { "dataPath": "params_shard_10.bin", "format": "raw-shard", "nbytes": 30240000, "records": [ { "name": "model.layers.12.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 0 }, { "name": "model.layers.12.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 4478976 }, { "name": "model.layers.12.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 5038848 }, { "name": "model.layers.12.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 6531840 }, { "name": "model.layers.12.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 6718464 }, { "name": "model.layers.12.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 16671744 }, { "name": "model.layers.12.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 17915904 }, { "name": "model.layers.12.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 22892544 }, { "name": "model.layers.12.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23514624 }, { "name": "model.layers.12.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23518080 }, { "name": "model.layers.13.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 23521536 }, { "name": "model.layers.13.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 28000512 }, { "name": "model.layers.13.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 28560384 }, { "name": "model.layers.13.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 30053376 } ], "md5sum": "0b8c6dfa6687c3db99e65117f79ba93d" }, { "dataPath": "params_shard_11.bin", "format": "raw-shard", "nbytes": 33474816, "records": [ { "name": "model.layers.13.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 0 }, { "name": "model.layers.13.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 9953280 }, { "name": "model.layers.13.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 11197440 }, { "name": "model.layers.13.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 16174080 }, { "name": "model.layers.13.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16796160 }, { "name": "model.layers.13.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16799616 }, { "name": "model.layers.14.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 16803072 }, { "name": "model.layers.14.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 21282048 }, { "name": "model.layers.14.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 21841920 }, { "name": "model.layers.14.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 23334912 }, { "name": "model.layers.14.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 23521536 } ], "md5sum": "2fbb1b71a2e438a1aac24f6118815035" }, { "dataPath": "params_shard_12.bin", "format": "raw-shard", "nbytes": 30371328, "records": [ { "name": "model.layers.14.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 0 }, { "name": "model.layers.14.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 1244160 }, { "name": "model.layers.14.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 6220800 }, { "name": "model.layers.14.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6842880 }, { "name": "model.layers.14.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6846336 }, { "name": "model.layers.15.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 6849792 }, { "name": "model.layers.15.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 11328768 }, { "name": "model.layers.15.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 11888640 }, { "name": "model.layers.15.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 13381632 }, { "name": "model.layers.15.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 13568256 }, { "name": "model.layers.15.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 23521536 }, { "name": "model.layers.15.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 24765696 }, { "name": "model.layers.15.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 29742336 }, { "name": "model.layers.15.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30364416 }, { "name": "model.layers.15.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30367872 } ], "md5sum": "4c4bec08b0a66fd02a837347c0f344da" }, { "dataPath": "params_shard_13.bin", "format": "raw-shard", "nbytes": 30240000, "records": [ { "name": "model.layers.16.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 0 }, { "name": "model.layers.16.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 4478976 }, { "name": "model.layers.16.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 5038848 }, { "name": "model.layers.16.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 6531840 }, { "name": "model.layers.16.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 6718464 }, { "name": "model.layers.16.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 16671744 }, { "name": "model.layers.16.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 17915904 }, { "name": "model.layers.16.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 22892544 }, { "name": "model.layers.16.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23514624 }, { "name": "model.layers.16.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23518080 }, { "name": "model.layers.17.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 23521536 }, { "name": "model.layers.17.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 28000512 }, { "name": "model.layers.17.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 28560384 }, { "name": "model.layers.17.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 30053376 } ], "md5sum": "57ff2396c7e7ff7a1759d5017f478bef" }, { "dataPath": "params_shard_14.bin", "format": "raw-shard", "nbytes": 33474816, "records": [ { "name": "model.layers.17.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 0 }, { "name": "model.layers.17.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 9953280 }, { "name": "model.layers.17.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 11197440 }, { "name": "model.layers.17.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 16174080 }, { "name": "model.layers.17.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16796160 }, { "name": "model.layers.17.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16799616 }, { "name": "model.layers.18.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 16803072 }, { "name": "model.layers.18.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 21282048 }, { "name": "model.layers.18.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 21841920 }, { "name": "model.layers.18.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 23334912 }, { "name": "model.layers.18.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 23521536 } ], "md5sum": "21e847b660a4b9ba586bb887dddd248e" }, { "dataPath": "params_shard_15.bin", "format": "raw-shard", "nbytes": 30371328, "records": [ { "name": "model.layers.18.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 0 }, { "name": "model.layers.18.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 1244160 }, { "name": "model.layers.18.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 6220800 }, { "name": "model.layers.18.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6842880 }, { "name": "model.layers.18.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6846336 }, { "name": "model.layers.19.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 6849792 }, { "name": "model.layers.19.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 11328768 }, { "name": "model.layers.19.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 11888640 }, { "name": "model.layers.19.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 13381632 }, { "name": "model.layers.19.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 13568256 }, { "name": "model.layers.19.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 23521536 }, { "name": "model.layers.19.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 24765696 }, { "name": "model.layers.19.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 29742336 }, { "name": "model.layers.19.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30364416 }, { "name": "model.layers.19.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30367872 } ], "md5sum": "0377995bec04e41216894044eed4168c" }, { "dataPath": "params_shard_16.bin", "format": "raw-shard", "nbytes": 30240000, "records": [ { "name": "model.layers.20.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 0 }, { "name": "model.layers.20.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 4478976 }, { "name": "model.layers.20.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 5038848 }, { "name": "model.layers.20.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 6531840 }, { "name": "model.layers.20.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 6718464 }, { "name": "model.layers.20.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 16671744 }, { "name": "model.layers.20.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 17915904 }, { "name": "model.layers.20.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 22892544 }, { "name": "model.layers.20.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23514624 }, { "name": "model.layers.20.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23518080 }, { "name": "model.layers.21.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 23521536 }, { "name": "model.layers.21.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 28000512 }, { "name": "model.layers.21.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 28560384 }, { "name": "model.layers.21.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 30053376 } ], "md5sum": "08128c243414472009ed83391530f239" }, { "dataPath": "params_shard_17.bin", "format": "raw-shard", "nbytes": 33474816, "records": [ { "name": "model.layers.21.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 0 }, { "name": "model.layers.21.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 9953280 }, { "name": "model.layers.21.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 11197440 }, { "name": "model.layers.21.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 16174080 }, { "name": "model.layers.21.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16796160 }, { "name": "model.layers.21.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16799616 }, { "name": "model.layers.22.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 16803072 }, { "name": "model.layers.22.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 21282048 }, { "name": "model.layers.22.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 21841920 }, { "name": "model.layers.22.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 23334912 }, { "name": "model.layers.22.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 23521536 } ], "md5sum": "401672d843ffe7481dc320f80afd4e96" }, { "dataPath": "params_shard_18.bin", "format": "raw-shard", "nbytes": 30371328, "records": [ { "name": "model.layers.22.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 0 }, { "name": "model.layers.22.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 1244160 }, { "name": "model.layers.22.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 6220800 }, { "name": "model.layers.22.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6842880 }, { "name": "model.layers.22.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6846336 }, { "name": "model.layers.23.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 6849792 }, { "name": "model.layers.23.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 11328768 }, { "name": "model.layers.23.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 11888640 }, { "name": "model.layers.23.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 13381632 }, { "name": "model.layers.23.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 13568256 }, { "name": "model.layers.23.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 23521536 }, { "name": "model.layers.23.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 24765696 }, { "name": "model.layers.23.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 29742336 }, { "name": "model.layers.23.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30364416 }, { "name": "model.layers.23.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30367872 } ], "md5sum": "e469722b8c8b2b31d6dca876ce6ec558" }, { "dataPath": "params_shard_19.bin", "format": "raw-shard", "nbytes": 30240000, "records": [ { "name": "model.layers.24.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 0 }, { "name": "model.layers.24.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 4478976 }, { "name": "model.layers.24.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 5038848 }, { "name": "model.layers.24.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 6531840 }, { "name": "model.layers.24.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 6718464 }, { "name": "model.layers.24.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 16671744 }, { "name": "model.layers.24.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 17915904 }, { "name": "model.layers.24.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 22892544 }, { "name": "model.layers.24.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23514624 }, { "name": "model.layers.24.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23518080 }, { "name": "model.layers.25.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 23521536 }, { "name": "model.layers.25.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 28000512 }, { "name": "model.layers.25.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 28560384 }, { "name": "model.layers.25.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 30053376 } ], "md5sum": "16ece73b6dc283c28aaf231bd245d329" }, { "dataPath": "params_shard_20.bin", "format": "raw-shard", "nbytes": 33474816, "records": [ { "name": "model.layers.25.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 0 }, { "name": "model.layers.25.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 9953280 }, { "name": "model.layers.25.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 11197440 }, { "name": "model.layers.25.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 16174080 }, { "name": "model.layers.25.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16796160 }, { "name": "model.layers.25.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16799616 }, { "name": "model.layers.26.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 16803072 }, { "name": "model.layers.26.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 21282048 }, { "name": "model.layers.26.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 21841920 }, { "name": "model.layers.26.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 23334912 }, { "name": "model.layers.26.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 23521536 } ], "md5sum": "411175e05a78a2ecc9510e404ff00cc2" }, { "dataPath": "params_shard_21.bin", "format": "raw-shard", "nbytes": 30371328, "records": [ { "name": "model.layers.26.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 0 }, { "name": "model.layers.26.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 1244160 }, { "name": "model.layers.26.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 6220800 }, { "name": "model.layers.26.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6842880 }, { "name": "model.layers.26.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6846336 }, { "name": "model.layers.27.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 6849792 }, { "name": "model.layers.27.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 11328768 }, { "name": "model.layers.27.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 11888640 }, { "name": "model.layers.27.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 13381632 }, { "name": "model.layers.27.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 13568256 }, { "name": "model.layers.27.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 23521536 }, { "name": "model.layers.27.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 24765696 }, { "name": "model.layers.27.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 29742336 }, { "name": "model.layers.27.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30364416 }, { "name": "model.layers.27.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30367872 } ], "md5sum": "37d3a4c0241aab0dd677d5a1c182a76e" }, { "dataPath": "params_shard_22.bin", "format": "raw-shard", "nbytes": 30240000, "records": [ { "name": "model.layers.28.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 0 }, { "name": "model.layers.28.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 4478976 }, { "name": "model.layers.28.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 5038848 }, { "name": "model.layers.28.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 6531840 }, { "name": "model.layers.28.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 6718464 }, { "name": "model.layers.28.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 16671744 }, { "name": "model.layers.28.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 17915904 }, { "name": "model.layers.28.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 22892544 }, { "name": "model.layers.28.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23514624 }, { "name": "model.layers.28.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23518080 }, { "name": "model.layers.29.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 23521536 }, { "name": "model.layers.29.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 28000512 }, { "name": "model.layers.29.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 28560384 }, { "name": "model.layers.29.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 30053376 } ], "md5sum": "e0b0a6471f62bd1abfab8bc66ec94d89" }, { "dataPath": "params_shard_23.bin", "format": "raw-shard", "nbytes": 33474816, "records": [ { "name": "model.layers.29.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 0 }, { "name": "model.layers.29.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 9953280 }, { "name": "model.layers.29.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 11197440 }, { "name": "model.layers.29.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 16174080 }, { "name": "model.layers.29.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16796160 }, { "name": "model.layers.29.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16799616 }, { "name": "model.layers.30.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 16803072 }, { "name": "model.layers.30.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 21282048 }, { "name": "model.layers.30.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 21841920 }, { "name": "model.layers.30.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 23334912 }, { "name": "model.layers.30.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 23521536 } ], "md5sum": "e49584e48d8ea18377e97dffa97304af" }, { "dataPath": "params_shard_24.bin", "format": "raw-shard", "nbytes": 30371328, "records": [ { "name": "model.layers.30.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 0 }, { "name": "model.layers.30.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 1244160 }, { "name": "model.layers.30.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 6220800 }, { "name": "model.layers.30.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6842880 }, { "name": "model.layers.30.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6846336 }, { "name": "model.layers.31.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 6849792 }, { "name": "model.layers.31.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 11328768 }, { "name": "model.layers.31.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 11888640 }, { "name": "model.layers.31.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 13381632 }, { "name": "model.layers.31.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 13568256 }, { "name": "model.layers.31.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 23521536 }, { "name": "model.layers.31.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 24765696 }, { "name": "model.layers.31.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 29742336 }, { "name": "model.layers.31.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30364416 }, { "name": "model.layers.31.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30367872 } ], "md5sum": "de170b58224667b2fc09b6b8178f45a1" }, { "dataPath": "params_shard_25.bin", "format": "raw-shard", "nbytes": 30240000, "records": [ { "name": "model.layers.32.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 0 }, { "name": "model.layers.32.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 4478976 }, { "name": "model.layers.32.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 5038848 }, { "name": "model.layers.32.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 6531840 }, { "name": "model.layers.32.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 6718464 }, { "name": "model.layers.32.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 16671744 }, { "name": "model.layers.32.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 17915904 }, { "name": "model.layers.32.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 22892544 }, { "name": "model.layers.32.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23514624 }, { "name": "model.layers.32.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23518080 }, { "name": "model.layers.33.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 23521536 }, { "name": "model.layers.33.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 28000512 }, { "name": "model.layers.33.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 28560384 }, { "name": "model.layers.33.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 30053376 } ], "md5sum": "e3204fd1f0367d708c176701ff1694b7" }, { "dataPath": "params_shard_26.bin", "format": "raw-shard", "nbytes": 33474816, "records": [ { "name": "model.layers.33.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 0 }, { "name": "model.layers.33.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 9953280 }, { "name": "model.layers.33.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 11197440 }, { "name": "model.layers.33.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 16174080 }, { "name": "model.layers.33.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16796160 }, { "name": "model.layers.33.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16799616 }, { "name": "model.layers.34.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 16803072 }, { "name": "model.layers.34.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 21282048 }, { "name": "model.layers.34.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 21841920 }, { "name": "model.layers.34.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 23334912 }, { "name": "model.layers.34.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 23521536 } ], "md5sum": "62a2c20688c6db06e549457ae105201c" }, { "dataPath": "params_shard_27.bin", "format": "raw-shard", "nbytes": 30371328, "records": [ { "name": "model.layers.34.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 0 }, { "name": "model.layers.34.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 1244160 }, { "name": "model.layers.34.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 6220800 }, { "name": "model.layers.34.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6842880 }, { "name": "model.layers.34.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6846336 }, { "name": "model.layers.35.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 6849792 }, { "name": "model.layers.35.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 11328768 }, { "name": "model.layers.35.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 11888640 }, { "name": "model.layers.35.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 13381632 }, { "name": "model.layers.35.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 13568256 }, { "name": "model.layers.35.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 23521536 }, { "name": "model.layers.35.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 24765696 }, { "name": "model.layers.35.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 29742336 }, { "name": "model.layers.35.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30364416 }, { "name": "model.layers.35.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30367872 } ], "md5sum": "94235b4b8ce06e51bc53b708737c4dfc" }, { "dataPath": "params_shard_28.bin", "format": "raw-shard", "nbytes": 30240000, "records": [ { "name": "model.layers.36.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 0 }, { "name": "model.layers.36.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 4478976 }, { "name": "model.layers.36.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 5038848 }, { "name": "model.layers.36.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 6531840 }, { "name": "model.layers.36.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 6718464 }, { "name": "model.layers.36.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 16671744 }, { "name": "model.layers.36.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 17915904 }, { "name": "model.layers.36.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 22892544 }, { "name": "model.layers.36.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23514624 }, { "name": "model.layers.36.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 23518080 }, { "name": "model.layers.37.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 23521536 }, { "name": "model.layers.37.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 28000512 }, { "name": "model.layers.37.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 28560384 }, { "name": "model.layers.37.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 30053376 } ], "md5sum": "e77c469d7437e46f41f66d599bcbfe24" }, { "dataPath": "params_shard_29.bin", "format": "raw-shard", "nbytes": 33474816, "records": [ { "name": "model.layers.37.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 0 }, { "name": "model.layers.37.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 9953280 }, { "name": "model.layers.37.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 11197440 }, { "name": "model.layers.37.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 16174080 }, { "name": "model.layers.37.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16796160 }, { "name": "model.layers.37.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 16799616 }, { "name": "model.layers.38.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 16803072 }, { "name": "model.layers.38.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 21282048 }, { "name": "model.layers.38.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 21841920 }, { "name": "model.layers.38.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 23334912 }, { "name": "model.layers.38.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 23521536 } ], "md5sum": "248d3dc6cfb69793b2ed5dbd86f62463" }, { "dataPath": "params_shard_30.bin", "format": "raw-shard", "nbytes": 424258560, "records": [ { "name": "lm_head.weight", "shape": [ 122760, 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 424258560, "byteOffset": 0 } ], "md5sum": "23346816a9b0c0bced3914b8cce3575b" }, { "dataPath": "params_shard_31.bin", "format": "raw-shard", "nbytes": 30374784, "records": [ { "name": "model.layers.38.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 0 }, { "name": "model.layers.38.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 1244160 }, { "name": "model.layers.38.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 6220800 }, { "name": "model.layers.38.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6842880 }, { "name": "model.layers.38.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 6846336 }, { "name": "model.layers.39.self_attn.wqkv_pack.q_weight", "shape": [ 5184, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4478976, "byteOffset": 6849792 }, { "name": "model.layers.39.self_attn.wqkv_pack.q_scale", "shape": [ 5184, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 559872, "byteOffset": 11328768 }, { "name": "model.layers.39.self_attn.o_proj.q_weight", "shape": [ 1728, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 1492992, "byteOffset": 11888640 }, { "name": "model.layers.39.self_attn.o_proj.q_scale", "shape": [ 1728, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 186624, "byteOffset": 13381632 }, { "name": "model.layers.39.mlp.gate_up_proj.q_weight", "shape": [ 11520, 216 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 9953280, "byteOffset": 13568256 }, { "name": "model.layers.39.mlp.gate_up_proj.q_scale", "shape": [ 11520, 54 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 1244160, "byteOffset": 23521536 }, { "name": "model.layers.39.mlp.down_proj.q_weight", "shape": [ 1728, 720 ], "dtype": "uint32", "format": "f32-to-bf16", "nbytes": 4976640, "byteOffset": 24765696 }, { "name": "model.layers.39.mlp.down_proj.q_scale", "shape": [ 1728, 180 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 622080, "byteOffset": 29742336 }, { "name": "model.layers.39.input_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30364416 }, { "name": "model.layers.39.post_attention_layernorm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30367872 }, { "name": "model.norm.weight", "shape": [ 1728 ], "dtype": "float16", "format": "f32-to-bf16", "nbytes": 3456, "byteOffset": 30371328 } ], "md5sum": "66416fbd70f99366d786699f0a74634f" } ] }