| { | |
| "metadata": { | |
| "ParamSize": 147, | |
| "ParamBytes": 2690863104.0, | |
| "BitsPerParam": 16.0 | |
| }, | |
| "records": [ | |
| { | |
| "dataPath": "params_shard_0.bin", | |
| "format": "raw-shard", | |
| "nbytes": 131080192, | |
| "records": [ | |
| { | |
| "name": "lm_head.weight", | |
| "shape": [ | |
| 32002, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 131080192, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "9f9e289791e66b5f020ed3adfd9733c5" | |
| }, | |
| { | |
| "dataPath": "params_shard_1.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.23.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "4d8a064245fdf9d5527440d3758c80fa" | |
| }, | |
| { | |
| "dataPath": "params_shard_2.bin", | |
| "format": "raw-shard", | |
| "nbytes": 131080192, | |
| "records": [ | |
| { | |
| "name": "model.embed_tokens.weight", | |
| "shape": [ | |
| 32002, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 131080192, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "b1d54ac38b97b992d841f2d125696f6c" | |
| }, | |
| { | |
| "dataPath": "params_shard_3.bin", | |
| "format": "raw-shard", | |
| "nbytes": 22544384, | |
| "records": [ | |
| { | |
| "name": "model.layers.0.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "0c9b5631c8d0782487272c1c05821ea0" | |
| }, | |
| { | |
| "dataPath": "params_shard_4.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.0.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "68ffe5cdc7b36404f6d9c117b005502f" | |
| }, | |
| { | |
| "dataPath": "params_shard_5.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.0.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "a1ee14cff03dde5b33e30af7d575a190" | |
| }, | |
| { | |
| "dataPath": "params_shard_6.bin", | |
| "format": "raw-shard", | |
| "nbytes": 22544384, | |
| "records": [ | |
| { | |
| "name": "model.layers.1.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "f17d8f11d5eb78bcd28a2ab102bd35df" | |
| }, | |
| { | |
| "dataPath": "params_shard_7.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.1.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "808fc0d79c4eb2667490bbb583798d04" | |
| }, | |
| { | |
| "dataPath": "params_shard_8.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.1.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "16485b4ac2d9d5a30eca613c1714bb28" | |
| }, | |
| { | |
| "dataPath": "params_shard_9.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30961664, | |
| "records": [ | |
| { | |
| "name": "model.layers.23.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.23.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 4096 | |
| }, | |
| { | |
| "name": "model.layers.23.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 22548480 | |
| }, | |
| { | |
| "name": "model.norm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 22552576 | |
| }, | |
| { | |
| "name": "model.layers.0.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 22556672 | |
| }, | |
| { | |
| "name": "model.layers.0.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 22560768 | |
| }, | |
| { | |
| "name": "model.layers.0.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 22564864 | |
| }, | |
| { | |
| "name": "model.layers.1.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30953472 | |
| }, | |
| { | |
| "name": "model.layers.1.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30957568 | |
| } | |
| ], | |
| "md5sum": "6c67534699b30cdf34f3b290f6585246" | |
| }, | |
| { | |
| "dataPath": "params_shard_10.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.10.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "0d4c78a526d638be4cda8f99eda54c52" | |
| }, | |
| { | |
| "dataPath": "params_shard_11.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.10.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "2e96d783ac5815edb36271486b860b91" | |
| }, | |
| { | |
| "dataPath": "params_shard_12.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.1.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.10.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.10.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.10.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "debfe6d51b8b897fb7bc6d96711c3ed3" | |
| }, | |
| { | |
| "dataPath": "params_shard_13.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.11.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "50ebc40a05f4aa4ce9859316ff83d9f8" | |
| }, | |
| { | |
| "dataPath": "params_shard_14.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.11.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "88a35ca8dbbee98cb90ffe2d61b9cde7" | |
| }, | |
| { | |
| "dataPath": "params_shard_15.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.10.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.11.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.11.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.11.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "fabe4ecbf48146653f53514f88857355" | |
| }, | |
| { | |
| "dataPath": "params_shard_16.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.12.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "b904d3951fbdc1591eef73a5c5efbbcd" | |
| }, | |
| { | |
| "dataPath": "params_shard_17.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.12.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "e84ed8d0bd39c0c4ea3a7bcf00b63225" | |
| }, | |
| { | |
| "dataPath": "params_shard_18.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.11.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.12.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.12.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.12.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "96663ca6bb7ee0f2e61c6d04f642bc0f" | |
| }, | |
| { | |
| "dataPath": "params_shard_19.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.13.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "151fc8b33d617d1ffbe07e75627f125b" | |
| }, | |
| { | |
| "dataPath": "params_shard_20.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.13.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "bcd9e7cf587b92c83f0d1e975aa17239" | |
| }, | |
| { | |
| "dataPath": "params_shard_21.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.12.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.13.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.13.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.13.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "c0f8c65a6ce70d0c1032459932d18d67" | |
| }, | |
| { | |
| "dataPath": "params_shard_22.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.14.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "01e1115542ac1e8e8f2aae519d2218d8" | |
| }, | |
| { | |
| "dataPath": "params_shard_23.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.14.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "51cc23816c40b0947aeeacf5ba47c108" | |
| }, | |
| { | |
| "dataPath": "params_shard_24.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.13.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.14.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.14.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.14.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "0083f49fda74b8cf84e5f1d07a900e16" | |
| }, | |
| { | |
| "dataPath": "params_shard_25.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.15.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "f360dce9c6b14b96cd26abb58e826863" | |
| }, | |
| { | |
| "dataPath": "params_shard_26.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.15.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "f4028906532c1a77c81fea68325f7e37" | |
| }, | |
| { | |
| "dataPath": "params_shard_27.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.14.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.15.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.15.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.15.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "3385aa8af474da3ecab88532ab76d44c" | |
| }, | |
| { | |
| "dataPath": "params_shard_28.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.16.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "9fd3b31de601dd1ea655d707fcb2b878" | |
| }, | |
| { | |
| "dataPath": "params_shard_29.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.16.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "ad0a0681984c1e8443e83da0aa5042ba" | |
| }, | |
| { | |
| "dataPath": "params_shard_30.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.15.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.16.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.16.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.16.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "d870f7ff9671024da16435a5c3d2e0ae" | |
| }, | |
| { | |
| "dataPath": "params_shard_31.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.17.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "364070562dba92fe20668f0dd6c4def1" | |
| }, | |
| { | |
| "dataPath": "params_shard_32.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.17.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "21ae9345d50c77cd3e07e29d93b9efdf" | |
| }, | |
| { | |
| "dataPath": "params_shard_33.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.16.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.17.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.17.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.17.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "7aeca74e93851c74a624201dd5ba5359" | |
| }, | |
| { | |
| "dataPath": "params_shard_34.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.18.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "65bb02353f04662f5ab3224e8eba760b" | |
| }, | |
| { | |
| "dataPath": "params_shard_35.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.18.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "da313fb40a2faba7e1903cb6754d2372" | |
| }, | |
| { | |
| "dataPath": "params_shard_36.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.17.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.18.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.18.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.18.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "602f91df1048aa4d59f5780c5a150e4e" | |
| }, | |
| { | |
| "dataPath": "params_shard_37.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.19.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "cde127fed4c562ffc9dfdd94298759f4" | |
| }, | |
| { | |
| "dataPath": "params_shard_38.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.19.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "5ecb5726da144dd9250eae22ce08d827" | |
| }, | |
| { | |
| "dataPath": "params_shard_39.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.18.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.19.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.19.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.19.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "df1e9f697e210b8f3d94e01c7f7023f3" | |
| }, | |
| { | |
| "dataPath": "params_shard_40.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.2.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "fed9db092ed650ce0d1154f0b128988c" | |
| }, | |
| { | |
| "dataPath": "params_shard_41.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.2.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "59a4ecebda8af22d714dc8a6b8075628" | |
| }, | |
| { | |
| "dataPath": "params_shard_42.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.19.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.2.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.2.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.2.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "852a934145f89a5017127dd06a380a6f" | |
| }, | |
| { | |
| "dataPath": "params_shard_43.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.20.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "1f05fcb53169a1d8be2dc20e5354d611" | |
| }, | |
| { | |
| "dataPath": "params_shard_44.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.20.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "3bf75fc9f377cfc71e0eb90c3b10daab" | |
| }, | |
| { | |
| "dataPath": "params_shard_45.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.2.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.20.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.20.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.20.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "b3723705c01ac8e5e1dbc899d70698b8" | |
| }, | |
| { | |
| "dataPath": "params_shard_46.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.21.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "6abbc9e86dd1e4f2013ff489952f9de7" | |
| }, | |
| { | |
| "dataPath": "params_shard_47.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.21.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "99e93cb34c970339bf5efb6875e38cb8" | |
| }, | |
| { | |
| "dataPath": "params_shard_48.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.20.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.21.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.21.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.21.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "78b764fd070d1615ba558d131f9e25dd" | |
| }, | |
| { | |
| "dataPath": "params_shard_49.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.22.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "8b647020224bda42aa46b0f4fd0b46f7" | |
| }, | |
| { | |
| "dataPath": "params_shard_50.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.22.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "4ed27a77728826cb5511b9171463c217" | |
| }, | |
| { | |
| "dataPath": "params_shard_51.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.21.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.22.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.22.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.22.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "e251ff26cdcff77317f4b118b02c1497" | |
| }, | |
| { | |
| "dataPath": "params_shard_52.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.23.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "36d8ea7c07c8bfcb2f88fa1fefe8ed79" | |
| }, | |
| { | |
| "dataPath": "params_shard_53.bin", | |
| "format": "raw-shard", | |
| "nbytes": 22544384, | |
| "records": [ | |
| { | |
| "name": "model.layers.3.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "517a9d2d31433214bdf434c6aa8a37e3" | |
| }, | |
| { | |
| "dataPath": "params_shard_54.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.3.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "aea80b9e667db9e3a0807c6debe68f0c" | |
| }, | |
| { | |
| "dataPath": "params_shard_55.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.3.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "cf9f8d29467431927a7bca62874a159b" | |
| }, | |
| { | |
| "dataPath": "params_shard_56.bin", | |
| "format": "raw-shard", | |
| "nbytes": 22544384, | |
| "records": [ | |
| { | |
| "name": "model.layers.4.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "4137120515b1fa42514d52e38434234d" | |
| }, | |
| { | |
| "dataPath": "params_shard_57.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.4.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "a6f9f971dc59ea038c22f9b3ccff60d4" | |
| }, | |
| { | |
| "dataPath": "params_shard_58.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.4.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "7a312b77164fd642e73b51ce7324b18f" | |
| }, | |
| { | |
| "dataPath": "params_shard_59.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25182208, | |
| "records": [ | |
| { | |
| "name": "model.layers.22.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.23.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.3.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 16777216 | |
| }, | |
| { | |
| "name": "model.layers.3.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 16781312 | |
| }, | |
| { | |
| "name": "model.layers.3.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 16785408 | |
| }, | |
| { | |
| "name": "model.layers.4.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 25174016 | |
| }, | |
| { | |
| "name": "model.layers.4.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 25178112 | |
| } | |
| ], | |
| "md5sum": "79413c6425e3979dc2d54af1f5f2acec" | |
| }, | |
| { | |
| "dataPath": "params_shard_60.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.5.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "07385011bf5cca986261e5627cb01477" | |
| }, | |
| { | |
| "dataPath": "params_shard_61.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.5.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "c163a8740f3a2634c4270b92bb3e27b3" | |
| }, | |
| { | |
| "dataPath": "params_shard_62.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.4.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.5.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.5.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.5.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "98999f0e1836f1bb330be0b9988ac1d1" | |
| }, | |
| { | |
| "dataPath": "params_shard_63.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.6.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "1b6d7427c4de377e95d02c74e95b51a8" | |
| }, | |
| { | |
| "dataPath": "params_shard_64.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.6.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "a64e834919358eafcf4437d4564603d1" | |
| }, | |
| { | |
| "dataPath": "params_shard_65.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.5.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.6.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.6.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.6.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "c530bf3867b0c3bae88cca854b8dbffd" | |
| }, | |
| { | |
| "dataPath": "params_shard_66.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.7.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "15273c470f4d0bff79539b2ec3e6ea65" | |
| }, | |
| { | |
| "dataPath": "params_shard_67.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.7.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "80328d7651bec73dbc301cf4fce5aa48" | |
| }, | |
| { | |
| "dataPath": "params_shard_68.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.6.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.7.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.7.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.7.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "cd936b0ee314daf92bc39651145ab724" | |
| }, | |
| { | |
| "dataPath": "params_shard_69.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.8.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "cdf81fb402a8a9f124c3603af04f43ac" | |
| }, | |
| { | |
| "dataPath": "params_shard_70.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.8.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "e5c5b7875ae5b54e3db24128edc79228" | |
| }, | |
| { | |
| "dataPath": "params_shard_71.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.7.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.8.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.8.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.8.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "423e7b1d04b274a992b8cc537daf1b61" | |
| }, | |
| { | |
| "dataPath": "params_shard_72.bin", | |
| "format": "raw-shard", | |
| "nbytes": 45088768, | |
| "records": [ | |
| { | |
| "name": "model.layers.9.mlp.gate_up_proj.weight", | |
| "shape": [ | |
| 11008, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 45088768, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "ff294c8adb898fb366b14302216e2101" | |
| }, | |
| { | |
| "dataPath": "params_shard_73.bin", | |
| "format": "raw-shard", | |
| "nbytes": 25165824, | |
| "records": [ | |
| { | |
| "name": "model.layers.9.self_attn.qkv_proj.weight", | |
| "shape": [ | |
| 6144, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 25165824, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "43a3490bd4d0e3426886c924f9e692b7" | |
| }, | |
| { | |
| "dataPath": "params_shard_74.bin", | |
| "format": "raw-shard", | |
| "nbytes": 30941184, | |
| "records": [ | |
| { | |
| "name": "model.layers.8.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| }, | |
| { | |
| "name": "model.layers.9.input_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 8388608 | |
| }, | |
| { | |
| "name": "model.layers.9.mlp.down_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 5504 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 22544384, | |
| "byteOffset": 8392704 | |
| }, | |
| { | |
| "name": "model.layers.9.post_attention_layernorm.weight", | |
| "shape": [ | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 4096, | |
| "byteOffset": 30937088 | |
| } | |
| ], | |
| "md5sum": "a233477ab303a8d9c73d40c8a3424cba" | |
| }, | |
| { | |
| "dataPath": "params_shard_75.bin", | |
| "format": "raw-shard", | |
| "nbytes": 8388608, | |
| "records": [ | |
| { | |
| "name": "model.layers.9.self_attn.o_proj.weight", | |
| "shape": [ | |
| 2048, | |
| 2048 | |
| ], | |
| "dtype": "float16", | |
| "format": "f32-to-bf16", | |
| "nbytes": 8388608, | |
| "byteOffset": 0 | |
| } | |
| ], | |
| "md5sum": "dd6db153a5d469094783da0810151802" | |
| } | |
| ] | |
| } |