diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..52373fe24473b1aa44333d318f578ae6bf04b49b 100644 --- a/.gitattributes +++ b/.gitattributes @@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.zip filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/config.json b/config.json new file mode 100644 index 0000000000000000000000000000000000000000..8defcb2e280c825886f2f74b1e3fa864d5e0d736 --- /dev/null +++ b/config.json @@ -0,0 +1,29 @@ +{ + "apply_residual_connection_post_layernorm": false, + "attention_dropout": 0.0, + "attention_softmax_in_fp32": true, + "bias_dropout_fusion": true, + "bos_token_id": 1, + "architectures": [ + "BloomForCausalLM" + ], + "eos_token_id": 2, + "pad_token_id": 3, + "unk_token_id": 0, + "hidden_dropout": 0.0, + "initializer_range": 0.02, + "layer_norm_epsilon": 1e-05, + "masked_softmax_fusion": true, + "model_type": "bloom", + "n_embed": 1536, + "n_inner": null, + "n_layer": 24, + "num_attention_heads": 16, + "offset_alibi": 100, + "pretraining_tp": 1, + "skip_bias_add": true, + "skip_bias_add_qkv": false, + "transformers_version": "4.20.0", + "use_cache": true, + "vocab_size": 250880 +} \ No newline at end of file diff --git a/model_00001-of-00026.safetensors b/model_00001-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a1f9097ab590231e07e72d0e28adc51723b262d9 --- /dev/null +++ b/model_00001-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abc0ab1034e852fcc45a4507f8b219b801d1dc9b673ee6b165c6a3dd9a64bfd7 +size 770709840 diff --git a/model_00002-of-00026.safetensors b/model_00002-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cc1fd59dd38817a71ff65411bb5af69de0ddc21f --- /dev/null +++ b/model_00002-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:323f2f8d69c5b580b86af15b7ed9573b52e09dfd2ab16847a038d4707ce5f432 +size 56664264 diff --git a/model_00003-of-00026.safetensors b/model_00003-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..61e0a80745b2fe360fc17defdb0852a954d26487 --- /dev/null +++ b/model_00003-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1dcfcaff173b21b8b952647082bca4396e7903591e0b618b07a17158898dba9c +size 56664264 diff --git a/model_00004-of-00026.safetensors b/model_00004-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..33fafa27e9ea089b7b1e479f65a8edd19aa8225f --- /dev/null +++ b/model_00004-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:06bf96319c892e9a565d4cb5b57236417ce8739afd77ccd5cf3563e331f5285d +size 56664264 diff --git a/model_00005-of-00026.safetensors b/model_00005-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..150a1a167bf9b6d11cd77c7b4378f76d8f6d8bb1 --- /dev/null +++ b/model_00005-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:20af575bec75b179d21eb0a21d8425875c686ade46b9569c5a933ab423d26a24 +size 56664264 diff --git a/model_00006-of-00026.safetensors b/model_00006-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5c03de5d7a409ca4f4b3beeb612a0da9d7d7c901 --- /dev/null +++ b/model_00006-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:057c0bb0623aad3012507f115e2886f94045f9fb623f4fc48404214fe0cb869d +size 56664264 diff --git a/model_00007-of-00026.safetensors b/model_00007-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..6721801bb33c2d1d92e95cc1f548ee179f31a29a --- /dev/null +++ b/model_00007-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14b526681aaeb83424ef29e6259f5b4c0480cd49ea5c62d53354de10ce6e5cc8 +size 56664264 diff --git a/model_00008-of-00026.safetensors b/model_00008-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f1b12b0f4ae245a4424da7fb93951dfb19303d99 --- /dev/null +++ b/model_00008-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:16eee89753711e948077e1a0c9454083443eb92f33dc4509df43fde784c02bba +size 56664264 diff --git a/model_00009-of-00026.safetensors b/model_00009-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d2b6eb5150c3cc6d6077038f2dd15e4669e0ee65 --- /dev/null +++ b/model_00009-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f8bbc62f7508a071bcf43e66570cc19de077c37775b28daadf9599ebcb47cc8b +size 56664264 diff --git a/model_00010-of-00026.safetensors b/model_00010-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8ad64f8dbb797604321f188c337dfcc166c5f278 --- /dev/null +++ b/model_00010-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4f45d23326732b551e311f6687310d424ff42b5ad4dcd0298a29985fe744f44c +size 56664264 diff --git a/model_00011-of-00026.safetensors b/model_00011-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..16cf33b77d2b8a73e842e72bd073d4632af2f57b --- /dev/null +++ b/model_00011-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4f8977110f765ca7f04682c83ac89dea59392f5e91f24f86cda62ba2e1bb82a5 +size 56664264 diff --git a/model_00012-of-00026.safetensors b/model_00012-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a1c1e02a1eaa4b7e3a06527c0431fc5699cb1b0a --- /dev/null +++ b/model_00012-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7b73577163af281f8857a86918def71a003b5dd179ff7e0be3d299c84a087482 +size 56664280 diff --git a/model_00013-of-00026.safetensors b/model_00013-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cf00ea0f59a5eb0b6fab31832a1119fc4f77844b --- /dev/null +++ b/model_00013-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5580cb5bcf3c53e349ae21931dc592201c6b0051f4b6cc1eb96c300fa1a51bce +size 56664280 diff --git a/model_00014-of-00026.safetensors b/model_00014-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..22c3a82180642a89fb8fc8b92460d2ff3cab076f --- /dev/null +++ b/model_00014-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d01c363c580e39bbf4cbde676428f70871baf28ca7f22acf25293c8778c7b55 +size 56664280 diff --git a/model_00015-of-00026.safetensors b/model_00015-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..067633fe465744eef1b131d7a5e072f5b66cd133 --- /dev/null +++ b/model_00015-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:78d6d2970207027d526b5d317a42ea77ff48079a74fdce27c754cbd158ef7150 +size 56664280 diff --git a/model_00016-of-00026.safetensors b/model_00016-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..9000f3e67fda912bf1b50762d7ce2063d687a0ff --- /dev/null +++ b/model_00016-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b4c7c19e9065815f94616cdc0889eba750245b6173771c46e1295ba73b45438e +size 56664280 diff --git a/model_00017-of-00026.safetensors b/model_00017-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..94329b56b2f3607f556cba9fb20e7d8298742294 --- /dev/null +++ b/model_00017-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c3693be88cfec1478e8c5d88400c1f0df91a42515c59ace2be81028c3240b2c0 +size 56664280 diff --git a/model_00018-of-00026.safetensors b/model_00018-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d6113f4ae7117854b8ef0debfaad9ed06e3b18cf --- /dev/null +++ b/model_00018-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3ccc3537231c1ceb6e2e972a4c750e641470d5864d2e3e17b5fd6de669e8082c +size 56664280 diff --git a/model_00019-of-00026.safetensors b/model_00019-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..64c3e652ceb6708843747bb2abd607f28a17237c --- /dev/null +++ b/model_00019-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5063ad85e42b1973c2349505990ec8e411f940b4390ca6fa5a50a95e2c908473 +size 56664280 diff --git a/model_00020-of-00026.safetensors b/model_00020-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..80c4bbf0626fa015466ef3643562297c2601821b --- /dev/null +++ b/model_00020-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:51b04c8c1ab5dbdd649a2759d0b105cf10647a5d29b5cdd0bd2c9291cb742c41 +size 56664280 diff --git a/model_00021-of-00026.safetensors b/model_00021-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..fadbd45575c26bbba706fd49ddbd273a1e020e8b --- /dev/null +++ b/model_00021-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8db4720514c79736b008f9bc2ad0d23f9c6667bc624d0b117c55d5dad0de1adf +size 56664280 diff --git a/model_00022-of-00026.safetensors b/model_00022-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c35669fd73854a7e1b12bfa630412243633bb4b6 --- /dev/null +++ b/model_00022-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4b509449d355075e9097d8afef0ace85a7ae315ecb7a4427ba71aaf66e3d3dc9 +size 56664280 diff --git a/model_00023-of-00026.safetensors b/model_00023-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..3b870a97bc18241ff3ba4c7f8b867bc52ea4aa69 --- /dev/null +++ b/model_00023-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:82181963de98b7a255336ae3019fb75b3792fe2dd2f19ec86af820e712b383b5 +size 56664280 diff --git a/model_00024-of-00026.safetensors b/model_00024-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1a91a66b442f2b3773f74bbd766eab5586bedcdf --- /dev/null +++ b/model_00024-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9c26b0717933b68dd40086ccd9ce5cef930599d60ba81747cfc022ebc6a15db +size 56664280 diff --git a/model_00025-of-00026.safetensors b/model_00025-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..569b9647a771e708ebc3958fa28e7b9248f1a46f --- /dev/null +++ b/model_00025-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:970b69147eeeee56235695ba605d7718fc31c68ae8a32e57fcd16ef4e0081f75 +size 56664280 diff --git a/model_00026-of-00026.safetensors b/model_00026-of-00026.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..56889c86951a7f4978c2122ed056d89e19968beb --- /dev/null +++ b/model_00026-of-00026.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77c20964ae9e1b0c34f466a74252e18910d23413e9da69232f77139d98928ca1 +size 6328 diff --git a/pytorch_model.bin.index.json b/pytorch_model.bin.index.json new file mode 100644 index 0000000000000000000000000000000000000000..97e3e7a6f1989dd761d530f18d93faf8a53d6abd --- /dev/null +++ b/pytorch_model.bin.index.json @@ -0,0 +1,300 @@ +{ + "metadata": { + "total_size": 2130628608 + }, + "weight_map": { + "h.0.input_layernorm.bias": "pytorch_model_00002-of-00026.bin", + "h.0.input_layernorm.weight": "pytorch_model_00002-of-00026.bin", + "h.0.mlp.dense_4h_to_h.bias": "pytorch_model_00002-of-00026.bin", + "h.0.mlp.dense_4h_to_h.weight": "pytorch_model_00002-of-00026.bin", + "h.0.mlp.dense_h_to_4h.bias": "pytorch_model_00002-of-00026.bin", + "h.0.mlp.dense_h_to_4h.weight": "pytorch_model_00002-of-00026.bin", + "h.0.post_attention_layernorm.bias": "pytorch_model_00002-of-00026.bin", + "h.0.post_attention_layernorm.weight": "pytorch_model_00002-of-00026.bin", + "h.0.self_attention.dense.bias": "pytorch_model_00002-of-00026.bin", + "h.0.self_attention.dense.weight": "pytorch_model_00002-of-00026.bin", + "h.0.self_attention.query_key_value.bias": "pytorch_model_00002-of-00026.bin", + "h.0.self_attention.query_key_value.weight": "pytorch_model_00002-of-00026.bin", + "h.1.input_layernorm.bias": "pytorch_model_00003-of-00026.bin", + "h.1.input_layernorm.weight": "pytorch_model_00003-of-00026.bin", + "h.1.mlp.dense_4h_to_h.bias": "pytorch_model_00003-of-00026.bin", + "h.1.mlp.dense_4h_to_h.weight": "pytorch_model_00003-of-00026.bin", + "h.1.mlp.dense_h_to_4h.bias": "pytorch_model_00003-of-00026.bin", + "h.1.mlp.dense_h_to_4h.weight": "pytorch_model_00003-of-00026.bin", + "h.1.post_attention_layernorm.bias": "pytorch_model_00003-of-00026.bin", + "h.1.post_attention_layernorm.weight": "pytorch_model_00003-of-00026.bin", + "h.1.self_attention.dense.bias": "pytorch_model_00003-of-00026.bin", + "h.1.self_attention.dense.weight": "pytorch_model_00003-of-00026.bin", + "h.1.self_attention.query_key_value.bias": "pytorch_model_00003-of-00026.bin", + "h.1.self_attention.query_key_value.weight": "pytorch_model_00003-of-00026.bin", + "h.10.input_layernorm.bias": "pytorch_model_00012-of-00026.bin", + "h.10.input_layernorm.weight": "pytorch_model_00012-of-00026.bin", + "h.10.mlp.dense_4h_to_h.bias": "pytorch_model_00012-of-00026.bin", + "h.10.mlp.dense_4h_to_h.weight": "pytorch_model_00012-of-00026.bin", + "h.10.mlp.dense_h_to_4h.bias": "pytorch_model_00012-of-00026.bin", + "h.10.mlp.dense_h_to_4h.weight": "pytorch_model_00012-of-00026.bin", + "h.10.post_attention_layernorm.bias": "pytorch_model_00012-of-00026.bin", + "h.10.post_attention_layernorm.weight": "pytorch_model_00012-of-00026.bin", + "h.10.self_attention.dense.bias": "pytorch_model_00012-of-00026.bin", + "h.10.self_attention.dense.weight": "pytorch_model_00012-of-00026.bin", + "h.10.self_attention.query_key_value.bias": "pytorch_model_00012-of-00026.bin", + "h.10.self_attention.query_key_value.weight": "pytorch_model_00012-of-00026.bin", + "h.11.input_layernorm.bias": "pytorch_model_00013-of-00026.bin", + "h.11.input_layernorm.weight": "pytorch_model_00013-of-00026.bin", + "h.11.mlp.dense_4h_to_h.bias": "pytorch_model_00013-of-00026.bin", + "h.11.mlp.dense_4h_to_h.weight": "pytorch_model_00013-of-00026.bin", + "h.11.mlp.dense_h_to_4h.bias": "pytorch_model_00013-of-00026.bin", + "h.11.mlp.dense_h_to_4h.weight": "pytorch_model_00013-of-00026.bin", + "h.11.post_attention_layernorm.bias": "pytorch_model_00013-of-00026.bin", + "h.11.post_attention_layernorm.weight": "pytorch_model_00013-of-00026.bin", + "h.11.self_attention.dense.bias": "pytorch_model_00013-of-00026.bin", + "h.11.self_attention.dense.weight": "pytorch_model_00013-of-00026.bin", + "h.11.self_attention.query_key_value.bias": "pytorch_model_00013-of-00026.bin", + "h.11.self_attention.query_key_value.weight": "pytorch_model_00013-of-00026.bin", + "h.12.input_layernorm.bias": "pytorch_model_00014-of-00026.bin", + "h.12.input_layernorm.weight": "pytorch_model_00014-of-00026.bin", + "h.12.mlp.dense_4h_to_h.bias": "pytorch_model_00014-of-00026.bin", + "h.12.mlp.dense_4h_to_h.weight": "pytorch_model_00014-of-00026.bin", + "h.12.mlp.dense_h_to_4h.bias": "pytorch_model_00014-of-00026.bin", + "h.12.mlp.dense_h_to_4h.weight": "pytorch_model_00014-of-00026.bin", + "h.12.post_attention_layernorm.bias": "pytorch_model_00014-of-00026.bin", + "h.12.post_attention_layernorm.weight": "pytorch_model_00014-of-00026.bin", + "h.12.self_attention.dense.bias": "pytorch_model_00014-of-00026.bin", + "h.12.self_attention.dense.weight": "pytorch_model_00014-of-00026.bin", + "h.12.self_attention.query_key_value.bias": "pytorch_model_00014-of-00026.bin", + "h.12.self_attention.query_key_value.weight": "pytorch_model_00014-of-00026.bin", + "h.13.input_layernorm.bias": "pytorch_model_00015-of-00026.bin", + "h.13.input_layernorm.weight": "pytorch_model_00015-of-00026.bin", + "h.13.mlp.dense_4h_to_h.bias": "pytorch_model_00015-of-00026.bin", + "h.13.mlp.dense_4h_to_h.weight": "pytorch_model_00015-of-00026.bin", + "h.13.mlp.dense_h_to_4h.bias": "pytorch_model_00015-of-00026.bin", + "h.13.mlp.dense_h_to_4h.weight": "pytorch_model_00015-of-00026.bin", + "h.13.post_attention_layernorm.bias": "pytorch_model_00015-of-00026.bin", + "h.13.post_attention_layernorm.weight": "pytorch_model_00015-of-00026.bin", + "h.13.self_attention.dense.bias": "pytorch_model_00015-of-00026.bin", + "h.13.self_attention.dense.weight": "pytorch_model_00015-of-00026.bin", + "h.13.self_attention.query_key_value.bias": "pytorch_model_00015-of-00026.bin", + "h.13.self_attention.query_key_value.weight": "pytorch_model_00015-of-00026.bin", + "h.14.input_layernorm.bias": "pytorch_model_00016-of-00026.bin", + "h.14.input_layernorm.weight": "pytorch_model_00016-of-00026.bin", + "h.14.mlp.dense_4h_to_h.bias": "pytorch_model_00016-of-00026.bin", + "h.14.mlp.dense_4h_to_h.weight": "pytorch_model_00016-of-00026.bin", + "h.14.mlp.dense_h_to_4h.bias": "pytorch_model_00016-of-00026.bin", + "h.14.mlp.dense_h_to_4h.weight": "pytorch_model_00016-of-00026.bin", + "h.14.post_attention_layernorm.bias": "pytorch_model_00016-of-00026.bin", + "h.14.post_attention_layernorm.weight": "pytorch_model_00016-of-00026.bin", + "h.14.self_attention.dense.bias": "pytorch_model_00016-of-00026.bin", + "h.14.self_attention.dense.weight": "pytorch_model_00016-of-00026.bin", + "h.14.self_attention.query_key_value.bias": "pytorch_model_00016-of-00026.bin", + "h.14.self_attention.query_key_value.weight": "pytorch_model_00016-of-00026.bin", + "h.15.input_layernorm.bias": "pytorch_model_00017-of-00026.bin", + "h.15.input_layernorm.weight": "pytorch_model_00017-of-00026.bin", + "h.15.mlp.dense_4h_to_h.bias": "pytorch_model_00017-of-00026.bin", + "h.15.mlp.dense_4h_to_h.weight": "pytorch_model_00017-of-00026.bin", + "h.15.mlp.dense_h_to_4h.bias": "pytorch_model_00017-of-00026.bin", + "h.15.mlp.dense_h_to_4h.weight": "pytorch_model_00017-of-00026.bin", + "h.15.post_attention_layernorm.bias": "pytorch_model_00017-of-00026.bin", + "h.15.post_attention_layernorm.weight": "pytorch_model_00017-of-00026.bin", + "h.15.self_attention.dense.bias": "pytorch_model_00017-of-00026.bin", + "h.15.self_attention.dense.weight": "pytorch_model_00017-of-00026.bin", + "h.15.self_attention.query_key_value.bias": "pytorch_model_00017-of-00026.bin", + "h.15.self_attention.query_key_value.weight": "pytorch_model_00017-of-00026.bin", + "h.16.input_layernorm.bias": "pytorch_model_00018-of-00026.bin", + "h.16.input_layernorm.weight": "pytorch_model_00018-of-00026.bin", + "h.16.mlp.dense_4h_to_h.bias": "pytorch_model_00018-of-00026.bin", + "h.16.mlp.dense_4h_to_h.weight": "pytorch_model_00018-of-00026.bin", + "h.16.mlp.dense_h_to_4h.bias": "pytorch_model_00018-of-00026.bin", + "h.16.mlp.dense_h_to_4h.weight": "pytorch_model_00018-of-00026.bin", + "h.16.post_attention_layernorm.bias": "pytorch_model_00018-of-00026.bin", + "h.16.post_attention_layernorm.weight": "pytorch_model_00018-of-00026.bin", + "h.16.self_attention.dense.bias": "pytorch_model_00018-of-00026.bin", + "h.16.self_attention.dense.weight": "pytorch_model_00018-of-00026.bin", + "h.16.self_attention.query_key_value.bias": "pytorch_model_00018-of-00026.bin", + "h.16.self_attention.query_key_value.weight": "pytorch_model_00018-of-00026.bin", + "h.17.input_layernorm.bias": "pytorch_model_00019-of-00026.bin", + "h.17.input_layernorm.weight": "pytorch_model_00019-of-00026.bin", + "h.17.mlp.dense_4h_to_h.bias": "pytorch_model_00019-of-00026.bin", + "h.17.mlp.dense_4h_to_h.weight": "pytorch_model_00019-of-00026.bin", + "h.17.mlp.dense_h_to_4h.bias": "pytorch_model_00019-of-00026.bin", + "h.17.mlp.dense_h_to_4h.weight": "pytorch_model_00019-of-00026.bin", + "h.17.post_attention_layernorm.bias": "pytorch_model_00019-of-00026.bin", + "h.17.post_attention_layernorm.weight": "pytorch_model_00019-of-00026.bin", + "h.17.self_attention.dense.bias": "pytorch_model_00019-of-00026.bin", + "h.17.self_attention.dense.weight": "pytorch_model_00019-of-00026.bin", + "h.17.self_attention.query_key_value.bias": "pytorch_model_00019-of-00026.bin", + "h.17.self_attention.query_key_value.weight": "pytorch_model_00019-of-00026.bin", + "h.18.input_layernorm.bias": "pytorch_model_00020-of-00026.bin", + "h.18.input_layernorm.weight": "pytorch_model_00020-of-00026.bin", + "h.18.mlp.dense_4h_to_h.bias": "pytorch_model_00020-of-00026.bin", + "h.18.mlp.dense_4h_to_h.weight": "pytorch_model_00020-of-00026.bin", + "h.18.mlp.dense_h_to_4h.bias": "pytorch_model_00020-of-00026.bin", + "h.18.mlp.dense_h_to_4h.weight": "pytorch_model_00020-of-00026.bin", + "h.18.post_attention_layernorm.bias": "pytorch_model_00020-of-00026.bin", + "h.18.post_attention_layernorm.weight": "pytorch_model_00020-of-00026.bin", + "h.18.self_attention.dense.bias": "pytorch_model_00020-of-00026.bin", + "h.18.self_attention.dense.weight": "pytorch_model_00020-of-00026.bin", + "h.18.self_attention.query_key_value.bias": "pytorch_model_00020-of-00026.bin", + "h.18.self_attention.query_key_value.weight": "pytorch_model_00020-of-00026.bin", + "h.19.input_layernorm.bias": "pytorch_model_00021-of-00026.bin", + "h.19.input_layernorm.weight": "pytorch_model_00021-of-00026.bin", + "h.19.mlp.dense_4h_to_h.bias": "pytorch_model_00021-of-00026.bin", + "h.19.mlp.dense_4h_to_h.weight": "pytorch_model_00021-of-00026.bin", + "h.19.mlp.dense_h_to_4h.bias": "pytorch_model_00021-of-00026.bin", + "h.19.mlp.dense_h_to_4h.weight": "pytorch_model_00021-of-00026.bin", + "h.19.post_attention_layernorm.bias": "pytorch_model_00021-of-00026.bin", + "h.19.post_attention_layernorm.weight": "pytorch_model_00021-of-00026.bin", + "h.19.self_attention.dense.bias": "pytorch_model_00021-of-00026.bin", + "h.19.self_attention.dense.weight": "pytorch_model_00021-of-00026.bin", + "h.19.self_attention.query_key_value.bias": "pytorch_model_00021-of-00026.bin", + "h.19.self_attention.query_key_value.weight": "pytorch_model_00021-of-00026.bin", + "h.2.input_layernorm.bias": "pytorch_model_00004-of-00026.bin", + "h.2.input_layernorm.weight": "pytorch_model_00004-of-00026.bin", + "h.2.mlp.dense_4h_to_h.bias": "pytorch_model_00004-of-00026.bin", + "h.2.mlp.dense_4h_to_h.weight": "pytorch_model_00004-of-00026.bin", + "h.2.mlp.dense_h_to_4h.bias": "pytorch_model_00004-of-00026.bin", + "h.2.mlp.dense_h_to_4h.weight": "pytorch_model_00004-of-00026.bin", + "h.2.post_attention_layernorm.bias": "pytorch_model_00004-of-00026.bin", + "h.2.post_attention_layernorm.weight": "pytorch_model_00004-of-00026.bin", + "h.2.self_attention.dense.bias": "pytorch_model_00004-of-00026.bin", + "h.2.self_attention.dense.weight": "pytorch_model_00004-of-00026.bin", + "h.2.self_attention.query_key_value.bias": "pytorch_model_00004-of-00026.bin", + "h.2.self_attention.query_key_value.weight": "pytorch_model_00004-of-00026.bin", + "h.20.input_layernorm.bias": "pytorch_model_00022-of-00026.bin", + "h.20.input_layernorm.weight": "pytorch_model_00022-of-00026.bin", + "h.20.mlp.dense_4h_to_h.bias": "pytorch_model_00022-of-00026.bin", + "h.20.mlp.dense_4h_to_h.weight": "pytorch_model_00022-of-00026.bin", + "h.20.mlp.dense_h_to_4h.bias": "pytorch_model_00022-of-00026.bin", + "h.20.mlp.dense_h_to_4h.weight": "pytorch_model_00022-of-00026.bin", + "h.20.post_attention_layernorm.bias": "pytorch_model_00022-of-00026.bin", + "h.20.post_attention_layernorm.weight": "pytorch_model_00022-of-00026.bin", + "h.20.self_attention.dense.bias": "pytorch_model_00022-of-00026.bin", + "h.20.self_attention.dense.weight": "pytorch_model_00022-of-00026.bin", + "h.20.self_attention.query_key_value.bias": "pytorch_model_00022-of-00026.bin", + "h.20.self_attention.query_key_value.weight": "pytorch_model_00022-of-00026.bin", + "h.21.input_layernorm.bias": "pytorch_model_00023-of-00026.bin", + "h.21.input_layernorm.weight": "pytorch_model_00023-of-00026.bin", + "h.21.mlp.dense_4h_to_h.bias": "pytorch_model_00023-of-00026.bin", + "h.21.mlp.dense_4h_to_h.weight": "pytorch_model_00023-of-00026.bin", + "h.21.mlp.dense_h_to_4h.bias": "pytorch_model_00023-of-00026.bin", + "h.21.mlp.dense_h_to_4h.weight": "pytorch_model_00023-of-00026.bin", + "h.21.post_attention_layernorm.bias": "pytorch_model_00023-of-00026.bin", + "h.21.post_attention_layernorm.weight": "pytorch_model_00023-of-00026.bin", + "h.21.self_attention.dense.bias": "pytorch_model_00023-of-00026.bin", + "h.21.self_attention.dense.weight": "pytorch_model_00023-of-00026.bin", + "h.21.self_attention.query_key_value.bias": "pytorch_model_00023-of-00026.bin", + "h.21.self_attention.query_key_value.weight": "pytorch_model_00023-of-00026.bin", + "h.22.input_layernorm.bias": "pytorch_model_00024-of-00026.bin", + "h.22.input_layernorm.weight": "pytorch_model_00024-of-00026.bin", + "h.22.mlp.dense_4h_to_h.bias": "pytorch_model_00024-of-00026.bin", + "h.22.mlp.dense_4h_to_h.weight": "pytorch_model_00024-of-00026.bin", + "h.22.mlp.dense_h_to_4h.bias": "pytorch_model_00024-of-00026.bin", + "h.22.mlp.dense_h_to_4h.weight": "pytorch_model_00024-of-00026.bin", + "h.22.post_attention_layernorm.bias": "pytorch_model_00024-of-00026.bin", + "h.22.post_attention_layernorm.weight": "pytorch_model_00024-of-00026.bin", + "h.22.self_attention.dense.bias": "pytorch_model_00024-of-00026.bin", + "h.22.self_attention.dense.weight": "pytorch_model_00024-of-00026.bin", + "h.22.self_attention.query_key_value.bias": "pytorch_model_00024-of-00026.bin", + "h.22.self_attention.query_key_value.weight": "pytorch_model_00024-of-00026.bin", + "h.23.input_layernorm.bias": "pytorch_model_00025-of-00026.bin", + "h.23.input_layernorm.weight": "pytorch_model_00025-of-00026.bin", + "h.23.mlp.dense_4h_to_h.bias": "pytorch_model_00025-of-00026.bin", + "h.23.mlp.dense_4h_to_h.weight": "pytorch_model_00025-of-00026.bin", + "h.23.mlp.dense_h_to_4h.bias": "pytorch_model_00025-of-00026.bin", + "h.23.mlp.dense_h_to_4h.weight": "pytorch_model_00025-of-00026.bin", + "h.23.post_attention_layernorm.bias": "pytorch_model_00025-of-00026.bin", + "h.23.post_attention_layernorm.weight": "pytorch_model_00025-of-00026.bin", + "h.23.self_attention.dense.bias": "pytorch_model_00025-of-00026.bin", + "h.23.self_attention.dense.weight": "pytorch_model_00025-of-00026.bin", + "h.23.self_attention.query_key_value.bias": "pytorch_model_00025-of-00026.bin", + "h.23.self_attention.query_key_value.weight": "pytorch_model_00025-of-00026.bin", + "h.3.input_layernorm.bias": "pytorch_model_00005-of-00026.bin", + "h.3.input_layernorm.weight": "pytorch_model_00005-of-00026.bin", + "h.3.mlp.dense_4h_to_h.bias": "pytorch_model_00005-of-00026.bin", + "h.3.mlp.dense_4h_to_h.weight": "pytorch_model_00005-of-00026.bin", + "h.3.mlp.dense_h_to_4h.bias": "pytorch_model_00005-of-00026.bin", + "h.3.mlp.dense_h_to_4h.weight": "pytorch_model_00005-of-00026.bin", + "h.3.post_attention_layernorm.bias": "pytorch_model_00005-of-00026.bin", + "h.3.post_attention_layernorm.weight": "pytorch_model_00005-of-00026.bin", + "h.3.self_attention.dense.bias": "pytorch_model_00005-of-00026.bin", + "h.3.self_attention.dense.weight": "pytorch_model_00005-of-00026.bin", + "h.3.self_attention.query_key_value.bias": "pytorch_model_00005-of-00026.bin", + "h.3.self_attention.query_key_value.weight": "pytorch_model_00005-of-00026.bin", + "h.4.input_layernorm.bias": "pytorch_model_00006-of-00026.bin", + "h.4.input_layernorm.weight": "pytorch_model_00006-of-00026.bin", + "h.4.mlp.dense_4h_to_h.bias": "pytorch_model_00006-of-00026.bin", + "h.4.mlp.dense_4h_to_h.weight": "pytorch_model_00006-of-00026.bin", + "h.4.mlp.dense_h_to_4h.bias": "pytorch_model_00006-of-00026.bin", + "h.4.mlp.dense_h_to_4h.weight": "pytorch_model_00006-of-00026.bin", + "h.4.post_attention_layernorm.bias": "pytorch_model_00006-of-00026.bin", + "h.4.post_attention_layernorm.weight": "pytorch_model_00006-of-00026.bin", + "h.4.self_attention.dense.bias": "pytorch_model_00006-of-00026.bin", + "h.4.self_attention.dense.weight": "pytorch_model_00006-of-00026.bin", + "h.4.self_attention.query_key_value.bias": "pytorch_model_00006-of-00026.bin", + "h.4.self_attention.query_key_value.weight": "pytorch_model_00006-of-00026.bin", + "h.5.input_layernorm.bias": "pytorch_model_00007-of-00026.bin", + "h.5.input_layernorm.weight": "pytorch_model_00007-of-00026.bin", + "h.5.mlp.dense_4h_to_h.bias": "pytorch_model_00007-of-00026.bin", + "h.5.mlp.dense_4h_to_h.weight": "pytorch_model_00007-of-00026.bin", + "h.5.mlp.dense_h_to_4h.bias": "pytorch_model_00007-of-00026.bin", + "h.5.mlp.dense_h_to_4h.weight": "pytorch_model_00007-of-00026.bin", + "h.5.post_attention_layernorm.bias": "pytorch_model_00007-of-00026.bin", + "h.5.post_attention_layernorm.weight": "pytorch_model_00007-of-00026.bin", + "h.5.self_attention.dense.bias": "pytorch_model_00007-of-00026.bin", + "h.5.self_attention.dense.weight": "pytorch_model_00007-of-00026.bin", + "h.5.self_attention.query_key_value.bias": "pytorch_model_00007-of-00026.bin", + "h.5.self_attention.query_key_value.weight": "pytorch_model_00007-of-00026.bin", + "h.6.input_layernorm.bias": "pytorch_model_00008-of-00026.bin", + "h.6.input_layernorm.weight": "pytorch_model_00008-of-00026.bin", + "h.6.mlp.dense_4h_to_h.bias": "pytorch_model_00008-of-00026.bin", + "h.6.mlp.dense_4h_to_h.weight": "pytorch_model_00008-of-00026.bin", + "h.6.mlp.dense_h_to_4h.bias": "pytorch_model_00008-of-00026.bin", + "h.6.mlp.dense_h_to_4h.weight": "pytorch_model_00008-of-00026.bin", + "h.6.post_attention_layernorm.bias": "pytorch_model_00008-of-00026.bin", + "h.6.post_attention_layernorm.weight": "pytorch_model_00008-of-00026.bin", + "h.6.self_attention.dense.bias": "pytorch_model_00008-of-00026.bin", + "h.6.self_attention.dense.weight": "pytorch_model_00008-of-00026.bin", + "h.6.self_attention.query_key_value.bias": "pytorch_model_00008-of-00026.bin", + "h.6.self_attention.query_key_value.weight": "pytorch_model_00008-of-00026.bin", + "h.7.input_layernorm.bias": "pytorch_model_00009-of-00026.bin", + "h.7.input_layernorm.weight": "pytorch_model_00009-of-00026.bin", + "h.7.mlp.dense_4h_to_h.bias": "pytorch_model_00009-of-00026.bin", + "h.7.mlp.dense_4h_to_h.weight": "pytorch_model_00009-of-00026.bin", + "h.7.mlp.dense_h_to_4h.bias": "pytorch_model_00009-of-00026.bin", + "h.7.mlp.dense_h_to_4h.weight": "pytorch_model_00009-of-00026.bin", + "h.7.post_attention_layernorm.bias": "pytorch_model_00009-of-00026.bin", + "h.7.post_attention_layernorm.weight": "pytorch_model_00009-of-00026.bin", + "h.7.self_attention.dense.bias": "pytorch_model_00009-of-00026.bin", + "h.7.self_attention.dense.weight": "pytorch_model_00009-of-00026.bin", + "h.7.self_attention.query_key_value.bias": "pytorch_model_00009-of-00026.bin", + "h.7.self_attention.query_key_value.weight": "pytorch_model_00009-of-00026.bin", + "h.8.input_layernorm.bias": "pytorch_model_00010-of-00026.bin", + "h.8.input_layernorm.weight": "pytorch_model_00010-of-00026.bin", + "h.8.mlp.dense_4h_to_h.bias": "pytorch_model_00010-of-00026.bin", + "h.8.mlp.dense_4h_to_h.weight": "pytorch_model_00010-of-00026.bin", + "h.8.mlp.dense_h_to_4h.bias": "pytorch_model_00010-of-00026.bin", + "h.8.mlp.dense_h_to_4h.weight": "pytorch_model_00010-of-00026.bin", + "h.8.post_attention_layernorm.bias": "pytorch_model_00010-of-00026.bin", + "h.8.post_attention_layernorm.weight": "pytorch_model_00010-of-00026.bin", + "h.8.self_attention.dense.bias": "pytorch_model_00010-of-00026.bin", + "h.8.self_attention.dense.weight": "pytorch_model_00010-of-00026.bin", + "h.8.self_attention.query_key_value.bias": "pytorch_model_00010-of-00026.bin", + "h.8.self_attention.query_key_value.weight": "pytorch_model_00010-of-00026.bin", + "h.9.input_layernorm.bias": "pytorch_model_00011-of-00026.bin", + "h.9.input_layernorm.weight": "pytorch_model_00011-of-00026.bin", + "h.9.mlp.dense_4h_to_h.bias": "pytorch_model_00011-of-00026.bin", + "h.9.mlp.dense_4h_to_h.weight": "pytorch_model_00011-of-00026.bin", + "h.9.mlp.dense_h_to_4h.bias": "pytorch_model_00011-of-00026.bin", + "h.9.mlp.dense_h_to_4h.weight": "pytorch_model_00011-of-00026.bin", + "h.9.post_attention_layernorm.bias": "pytorch_model_00011-of-00026.bin", + "h.9.post_attention_layernorm.weight": "pytorch_model_00011-of-00026.bin", + "h.9.self_attention.dense.bias": "pytorch_model_00011-of-00026.bin", + "h.9.self_attention.dense.weight": "pytorch_model_00011-of-00026.bin", + "h.9.self_attention.query_key_value.bias": "pytorch_model_00011-of-00026.bin", + "h.9.self_attention.query_key_value.weight": "pytorch_model_00011-of-00026.bin", + "ln_f.bias": "pytorch_model_00026-of-00026.bin", + "ln_f.weight": "pytorch_model_00026-of-00026.bin", + "word_embeddings.weight": "pytorch_model_00001-of-00026.bin", + "word_embeddings_layernorm.bias": "pytorch_model_00001-of-00026.bin", + "word_embeddings_layernorm.weight": "pytorch_model_00001-of-00026.bin" + } +} diff --git a/pytorch_model_00001-of-00026.bin b/pytorch_model_00001-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..d2012e4fbadbeaf5fd091c265fa0b388deaed4da --- /dev/null +++ b/pytorch_model_00001-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc73b4a208404b01da1eabd2667a720b49b428534bfcb97fd403eab3980cf9d7 +size 770711527 diff --git a/pytorch_model_00002-of-00026.bin b/pytorch_model_00002-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..58d990cff6a9f7651b7e6f2271009ec98353341c --- /dev/null +++ b/pytorch_model_00002-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8a6fc1efc151ac025aaec657c02e05dd1e8683544c041fd6d2d6b441e17bf798 +size 56667778 diff --git a/pytorch_model_00003-of-00026.bin b/pytorch_model_00003-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..a370871a3a433ce15f2184f67c9ad48a4c84ea43 --- /dev/null +++ b/pytorch_model_00003-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d552aba533de65b990d9f01ad2e2eceb61959395174d0d4ddc53442b31ff995e +size 56667778 diff --git a/pytorch_model_00004-of-00026.bin b/pytorch_model_00004-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..7e56b9e2972511091483c4f994453d582fb9853b --- /dev/null +++ b/pytorch_model_00004-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:61d6e50b119e295c31589bd915cb240ada412bba1a116df8493b8cf761739d21 +size 56667778 diff --git a/pytorch_model_00005-of-00026.bin b/pytorch_model_00005-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..4ffbda604c16d4ee6576e4731ec497991cac8657 --- /dev/null +++ b/pytorch_model_00005-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:572ddfb49d0a81c591dafe05a5aa9e2e5eba6615302fa7d33f1c3ad610b40312 +size 56667778 diff --git a/pytorch_model_00006-of-00026.bin b/pytorch_model_00006-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..36d812fb41e3fe3c186ca41c196d9b377f52c4b6 --- /dev/null +++ b/pytorch_model_00006-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa9bfcdd71e0d9bb4b0b8d7cdb1a6e79b3938beaf39295f3bf259bc5a0745503 +size 56667778 diff --git a/pytorch_model_00007-of-00026.bin b/pytorch_model_00007-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..e60ef80d71ce60af176a28579a4c2e23e780730d --- /dev/null +++ b/pytorch_model_00007-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ddd2563300c1eb8210ada0c6c2ba5f43a253466a90a669501c194519116ac594 +size 56667778 diff --git a/pytorch_model_00008-of-00026.bin b/pytorch_model_00008-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..9d5cde813597c8afa78a6b975d3910dd4bbea5e7 --- /dev/null +++ b/pytorch_model_00008-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ddc95be16fed815e3a0591eab5a01483860d2c7494b750fe59826092badc2f5 +size 56667778 diff --git a/pytorch_model_00009-of-00026.bin b/pytorch_model_00009-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..d009eebf84ee96d92b2fdc6bf355194d09018e58 --- /dev/null +++ b/pytorch_model_00009-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4c8ae9baf01ae10b61dc0ce653833aac2ab70c64497ba40203beb538efcb6f25 +size 56667778 diff --git a/pytorch_model_00010-of-00026.bin b/pytorch_model_00010-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..4377a0d5597b962b3a1dd232bd34cba819dd76e2 --- /dev/null +++ b/pytorch_model_00010-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a37b714fd5137aeb8743960866c2ce688d2ec92d8f8ff781611ffbcb84691ae +size 56667778 diff --git a/pytorch_model_00011-of-00026.bin b/pytorch_model_00011-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..76d975207f42f49d6946e4c2e0ee6d7eb4a459d9 --- /dev/null +++ b/pytorch_model_00011-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a13bc46d2b9dcb86137081651359bcdcf9f48bc1fa04ed30cc7fce80090034b6 +size 56667778 diff --git a/pytorch_model_00012-of-00026.bin b/pytorch_model_00012-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..567c53109c7b0660614716f69b4957efe74e2df2 --- /dev/null +++ b/pytorch_model_00012-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0cad596338757ab5ec4be1171767e34edba24855f48638c4c61e1924f4e78c85 +size 56667778 diff --git a/pytorch_model_00013-of-00026.bin b/pytorch_model_00013-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..a63b76aab71e8cfcd6495eb464f522284aee9b8e --- /dev/null +++ b/pytorch_model_00013-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee19c6e7365e09d8457cb7257ce48c3dac2fbbd21dddb83f11598425e1fbf5e5 +size 56667778 diff --git a/pytorch_model_00014-of-00026.bin b/pytorch_model_00014-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..a767eb442b5c7466597754206c1e5d600b208187 --- /dev/null +++ b/pytorch_model_00014-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:89336fd6e6fd59c03dec6754d068833c0f2987383923930e29c7e88c3f7d0a06 +size 56667778 diff --git a/pytorch_model_00015-of-00026.bin b/pytorch_model_00015-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..2470d51b0f52eca25c8b6873143a9ae25a4cf288 --- /dev/null +++ b/pytorch_model_00015-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa5eceff01c1cea0f00de992b4776da5b68101078c9a75fc7c63117a8d0226f7 +size 56667778 diff --git a/pytorch_model_00016-of-00026.bin b/pytorch_model_00016-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..2de66255afc46544969be60b4159dbd019ffa8ab --- /dev/null +++ b/pytorch_model_00016-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:80dff1d94ae89e7e4a3e32cf508c5a152206575a5968e3557d60e2b9807fb9fa +size 56667778 diff --git a/pytorch_model_00017-of-00026.bin b/pytorch_model_00017-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..8aa3c7cd4344d928d5a52630ad4b8a4660e8efb9 --- /dev/null +++ b/pytorch_model_00017-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c45489c0a48371e4e90dc8d1bae9da0a3e92bde10c03c5413f4e1e4ef293a83 +size 56667778 diff --git a/pytorch_model_00018-of-00026.bin b/pytorch_model_00018-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..31b8f191f69cab35926c903f332a844f2cea979a --- /dev/null +++ b/pytorch_model_00018-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9a74987e4e7d9c3a91f3153585de04bfececa0539851b4c49cfdc7558566722 +size 56667778 diff --git a/pytorch_model_00019-of-00026.bin b/pytorch_model_00019-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..4c11779558032c8b3b615b72e7a426e7b4c88349 --- /dev/null +++ b/pytorch_model_00019-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef27ad2ac0fde596e992901326302438870b7c3334028ba1d30f46a39ccdf9df +size 56667778 diff --git a/pytorch_model_00020-of-00026.bin b/pytorch_model_00020-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..c09c7c422d8d2c21645a754cfa2fbb6e2eb3751e --- /dev/null +++ b/pytorch_model_00020-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f17ae508ca43c55f83ba14fab1d620f5f0286f5b909b3f4f1f082c934e12a15d +size 56667778 diff --git a/pytorch_model_00021-of-00026.bin b/pytorch_model_00021-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..2a5eea9ba6bbf721e1a781bb307e8591f9116f79 --- /dev/null +++ b/pytorch_model_00021-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f327c005f4f49ad35460aac760ea0908211480254e20206a8836720aeef709e6 +size 56667778 diff --git a/pytorch_model_00022-of-00026.bin b/pytorch_model_00022-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..dcb2877e4911ba18629ac9ac7d95b6f4fd61c0ad --- /dev/null +++ b/pytorch_model_00022-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:70725e2d7269afd0dcbe9f7a8f696cf93465a812946bae7465756dcebfc6de22 +size 56667778 diff --git a/pytorch_model_00023-of-00026.bin b/pytorch_model_00023-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..7718ebc7f01f4e0ae3a4ad32b5c715f0ba1207fe --- /dev/null +++ b/pytorch_model_00023-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:24362688e52988ccbbdef151865b8e2a3fd80fffe312c6ed21b99baf1183c32a +size 56667778 diff --git a/pytorch_model_00024-of-00026.bin b/pytorch_model_00024-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..5b6ed931698cc3f313f4e30dceef696f9a4d8b6e --- /dev/null +++ b/pytorch_model_00024-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ba13d10b8afdf84bf0d01609332666d5ecf6cbccd17dad34f73aed59e4876f2b +size 56667778 diff --git a/pytorch_model_00025-of-00026.bin b/pytorch_model_00025-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..b97897e76b9bdcbcaab005fecdea7ec0ba6771b5 --- /dev/null +++ b/pytorch_model_00025-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cc8bafb55afd3717afb03557bea62eda966479654981fb17a9f712e30acc778a +size 56667778 diff --git a/pytorch_model_00026-of-00026.bin b/pytorch_model_00026-of-00026.bin new file mode 100644 index 0000000000000000000000000000000000000000..55acf36ea9ce428e961718739798243ce97b677b --- /dev/null +++ b/pytorch_model_00026-of-00026.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e4aa4cf1f34e3564ac687e11b4920a1b6e685bb4ab97fd70f66e3564c4cda77f +size 7830 diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..25bc39604f72700b3b8e10bd69bb2f227157edd1 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1 @@ +{"bos_token": "", "eos_token": "", "unk_token": "", "pad_token": ""} \ No newline at end of file diff --git a/text_embed_service.py b/text_embed_service.py new file mode 100644 index 0000000000000000000000000000000000000000..327b299134efdfea90ad3794d78167c8d540b26d --- /dev/null +++ b/text_embed_service.py @@ -0,0 +1,111 @@ +import torch +import cherrypy +import cherrypy_cors +from argparse import ArgumentParser +from transformers import BertTokenizer, BertModel, AutoTokenizer, AutoModel +from sentence_transformers import SentenceTransformer + + +class MyWebService(object): + @cherrypy.expose + @cherrypy.tools.json_out() + @cherrypy.tools.json_in() + def index(self): + try: + data = cherrypy.request.json + except TypeError: + return {'error': 'invalid input'} + representation = calculate_embedding(data['input']) + return {'embedding': representation.tolist()} + + +def mean_pooling(model_output, attention_mask): + token_embeddings = model_output[0] # First element of model_output contains all token embeddings + input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() + return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) + + +def arg_loader(): + parser = ArgumentParser() + parser.add_argument('--model', type=str, default='text2vec_large') + parser.add_argument('--port', type=int, default=8500) + parser.add_argument('--gpu', type=int, default=0) + parser.add_argument('--batch_size', type=int, default=128) + return parser.parse_args() + + +def calculate_embedding(sentences): + if args.model in ['text2vec_large']: + result = [] + model.eval() + with torch.no_grad(): + i = 0 + while i < len(sentences): + tokens = tokenizer(sentences[i:i + args.batch_size], padding=True, truncation=True, max_length=512, + return_tensors='pt') + if torch.cuda.is_available(): + tokens = {k: v.cuda(args.gpu) for k, v in tokens.items()} + outputs = model(**tokens) + + if args.model == 'text2vec_large': + outputs = mean_pooling(outputs, tokens['attention_mask']) + # elif args.model == 'gte_base': + # outputs = outputs.last_hidden_state[:, 0] + else: + raise NotImplementedError + + outputs = torch.nn.functional.normalize(outputs) + result.append(outputs) + i += args.batch_size + result = torch.cat(result, dim=0) + else: + result = model.encode(sentences, batch_size=args.batch_size, normalize_embeddings=True) + return result + + +if __name__ == '__main__': + args = arg_loader() + print('================') + for key, value in vars(args).items(): + print(key, value) + print('================') + + if args.model == 'text2vec_large': + tokenizer = BertTokenizer.from_pretrained('GanymedeNil/text2vec-large-chinese') + model = BertModel.from_pretrained('GanymedeNil/text2vec-large-chinese') + elif args.model == 'gte_base': + model = SentenceTransformer('thenlper/gte-base-zh') + elif args.model == 'gte_large': + model = SentenceTransformer('thenlper/gte-large-zh') + elif args.model == 'zpoint_large': + model = SentenceTransformer('iampanda/zpoint_large_embedding_zh') + elif args.model == 'yinka': + model = SentenceTransformer('Classical/Yinka') + elif args.model == 'acge': + model = SentenceTransformer('aspire/acge_text_embedding') + elif args.model == 'stella_mrl_large': + model = SentenceTransformer('infgrad/stella-mrl-large-zh-v3.5-1792d') + elif args.model == 'stella_large': + model = SentenceTransformer('infgrad/stella-large-zh-v3-1792d') + elif args.model == 'gte_large_en': + model = SentenceTransformer('Alibaba-NLP/gte-large-en-v1.5', trust_remote_code=True) + else: + raise NotImplementedError + + if torch.cuda.is_available(): + print('moving model to gpu %d' % args.gpu) + model = model.cuda(args.gpu) + + print('starting service...') + cherrypy_cors.install() + config = { + 'global': { + 'server.socket_host': '0.0.0.0', + 'server.socket_port': args.port, + 'cors.expose.on': True + } + } + cherrypy.config.update(config) + cherrypy.quickstart(MyWebService(), '/', config) + +# python3 service.py --model zpoint_large --gpu 0 --port 8500 --batch_size 128 diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..370bd68e20b4b6574ee05b213a74b244e3f492f3 --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3fa39cd4b1500feb205bcce3b9703a4373414cafe4970e0657b413f7ddd2a9d3 +size 14500438 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..1cb46ea45a11a1992a06c86be4e7f8132a17c17c --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1 @@ +{"unk_token": "", "eos_token": "", "bos_token": "", "pad_token": "", "name_or_path": "bigscience/tokenizer", "special_tokens_map_file": null, "tokenizer_class": "PreTrainedTokenizerFast", "padding_side":"left"} \ No newline at end of file