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..73e4e2f18738034d1678b68bf3bde335a0116057 --- /dev/null +++ b/config.json @@ -0,0 +1,38 @@ +{ + "architectures": [ + "MixtralForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 1, + "dtype": "bfloat16", + "eos_token_id": 2, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 4096, + "mlp_bias": false, + "model_type": "mixtral", + "num_attention_heads": 32, + "num_experts_per_tok": 2, + "num_hidden_layers": 48, + "num_key_value_heads": 8, + "num_local_experts": 2, + "output_router_logits": false, + "pad_token_id": 2, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_parameters": { + "rope_theta": 10000.0, + "rope_type": "default" + }, + "router_aux_loss_coef": 0.001, + "router_jitter_noise": 0.0, + "sliding_window": null, + "tie_word_embeddings": false, + "transformers_version": "5.10.1", + "use_cache": true, + "vocab_size": 262162 +} \ No newline at end of file diff --git a/model-00001-of-00081.safetensors b/model-00001-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..67c38d37d0f7db56d0990ede6895b7157fef3d21 --- /dev/null +++ b/model-00001-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa86d519f5dbd8db584d8881354927436166eda369c7efae56366918a7e79b99 +size 2348975008 diff --git a/model-00002-of-00081.safetensors b/model-00002-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..117dcee3f35409f0601b29b844f287116cf5b3f4 --- /dev/null +++ b/model-00002-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5038311075e42dba835f9fa27bfc35ce41db7c7f5889a804854f60ed48c8bd1 +size 469762600 diff --git a/model-00003-of-00081.safetensors b/model-00003-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..01815e60a3c0830c2195c11fdd061f0e0b12230c --- /dev/null +++ b/model-00003-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c3b0a5375b518c029b668bedfdc30259f97ff83dcb83a003174ed65fd5746382 +size 436225112 diff --git a/model-00004-of-00081.safetensors b/model-00004-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f7c0d6523a670caf97d0017450fe787b94ff71f5 --- /dev/null +++ b/model-00004-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:266c4eb5e6c5aec934cdf14b156942bc974d18dfbb36d48b169c747c6ec3f0be +size 520102896 diff --git a/model-00005-of-00081.safetensors b/model-00005-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..9f62b99d26f3193058d77b4f0d50d39a18079323 --- /dev/null +++ b/model-00005-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5bca5eec75c47460b28b812859522c9cff9704834f4f3a2ecee4a834b67b32aa +size 503325456 diff --git a/model-00006-of-00081.safetensors b/model-00006-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a53a83bdfb9e40caa53bf5dabd003552b58e8e62 --- /dev/null +++ b/model-00006-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:547e45015c56d496c5d119010d9a61e61997457eab96a81177db63140cb6f5a9 +size 436225112 diff --git a/model-00007-of-00081.safetensors b/model-00007-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ce87180f930b31761cc4a3e7b7e5999a02093786 --- /dev/null +++ b/model-00007-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee49e768858d6ac691dd8f135373fe2994723a4ffddd4278ca07f409ce36e76f +size 469762600 diff --git a/model-00008-of-00081.safetensors b/model-00008-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b4e83eba158134d6deec006ce30c6422c7f4711f --- /dev/null +++ b/model-00008-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:96490e7deef72dfbcb8682dcd8d25f9a5a4f24d3d56d129a55d97ec5f91ff80f +size 436225112 diff --git a/model-00009-of-00081.safetensors b/model-00009-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..efd882f6f6cde5051451b93ba0f9b3174f9c96f1 --- /dev/null +++ b/model-00009-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:260770aac94b15d49b8cf7553beab2f5498e0ff3bbbf758ce58e91591c1b29c5 +size 520102896 diff --git a/model-00010-of-00081.safetensors b/model-00010-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..fd047b1d998e8f32b657016c228074a1ccc90c36 --- /dev/null +++ b/model-00010-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b1a8ecb04a94f63b5c76391b6d364ad4294f0ca6e7ca4040c7d22a71e585cf38 +size 503325456 diff --git a/model-00011-of-00081.safetensors b/model-00011-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b01318add7504cffe160fec894a860d2e8a4d7e6 --- /dev/null +++ b/model-00011-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:681ed2f427adbd4329fc69d952cda617538e27badd1b806c6a2458bc0da07447 +size 436225112 diff --git a/model-00012-of-00081.safetensors b/model-00012-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..61be1522b0fb8d3a9f2e80188ecd325879d8dc26 --- /dev/null +++ b/model-00012-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e2e4524e64f932b5bce867c69f59f98c5504094f2a8872d3d282bf6ca77a7ab2 +size 469762600 diff --git a/model-00013-of-00081.safetensors b/model-00013-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..78468f41c7c960b5d35e361e9ca19086a6e5e5de --- /dev/null +++ b/model-00013-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:41180db84ccadf048c4bd59f51ac4fcd716c457d52fb892e9e852a7987dfd2f5 +size 436225112 diff --git a/model-00014-of-00081.safetensors b/model-00014-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1831d0cd127528f9a3cb6773b29fb5a9679a6f6e --- /dev/null +++ b/model-00014-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1a492f600203e8fafaa2976aa12f7ed05961efa4c0572b77cf29fa2ca7cd0478 +size 520102896 diff --git a/model-00015-of-00081.safetensors b/model-00015-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f92f2f0e07f1d8cdc7e630a73357107ef20afe5c --- /dev/null +++ b/model-00015-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b06fb5a5c0528f33a14e2597642ab7622ec30b29ba2c10bcfc545b75178415ce +size 503325456 diff --git a/model-00016-of-00081.safetensors b/model-00016-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a4d47c15e511613143c00e41a9c67b8edf8a7392 --- /dev/null +++ b/model-00016-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:da132d174119718f27e480bd2db3cefc04d0392837f4a8ed985d27d9ed6ec033 +size 436225112 diff --git a/model-00017-of-00081.safetensors b/model-00017-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..52e509c5fadb6e14dfc462622f01c2dc00cbd174 --- /dev/null +++ b/model-00017-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0529db14e9c771f4ff129f1fe889c7f71963d2e07877ea17958b337c9a053035 +size 469762600 diff --git a/model-00018-of-00081.safetensors b/model-00018-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..6c4e2e4de5d7f7c541809a4678cd0de62c530c03 --- /dev/null +++ b/model-00018-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c0cca9528f47f2a0731583628f036718d844e1abc53f96f6e7fbc6210540aef +size 436225120 diff --git a/model-00019-of-00081.safetensors b/model-00019-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..32f30c326a524333722b94bbffadd7ed3501947a --- /dev/null +++ b/model-00019-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:593cd7aebab16b25ae2e38a35b2ad0d1b4d4638435983b6cee58e5cc5cdf4361 +size 520102904 diff --git a/model-00020-of-00081.safetensors b/model-00020-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c6ad82f1c14b1a3519e25da5387e6c56b1f63284 --- /dev/null +++ b/model-00020-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:54b044295b067a8b0c98b9c2ded6a4d7684a722b82cc01edbb736529f5df8e2f +size 503325464 diff --git a/model-00021-of-00081.safetensors b/model-00021-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..31640c147395db9129aee4ad22104bc9512142c2 --- /dev/null +++ b/model-00021-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1c1ec091fba01367ac0c96cca4f45d4ef0d1001468a7963bc9d0dbea379b3c57 +size 436225120 diff --git a/model-00022-of-00081.safetensors b/model-00022-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d500cd5a4ccf9ffbc0b231aa2023300bc5f39128 --- /dev/null +++ b/model-00022-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:65774d77f4c92b24d00950ff0abf0f1fefa8e2907eb56f9a63328175b5bbed74 +size 469762600 diff --git a/model-00023-of-00081.safetensors b/model-00023-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a91c6b4f407de68fa1ab0c1900b50765113f0c68 --- /dev/null +++ b/model-00023-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:94293cd03a1a644f399ab24fcb684c0ae9b86e05e0648631fc0399a971938a70 +size 436225120 diff --git a/model-00024-of-00081.safetensors b/model-00024-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cb7e927c8c4d4f1d690b29ca980b62feab3592a0 --- /dev/null +++ b/model-00024-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30d54b78caa08af8921867233aacae6ca872d957c268f215be03051857f32a09 +size 520102904 diff --git a/model-00025-of-00081.safetensors b/model-00025-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..59761580b380b94964757d48d109d34feb9dd37d --- /dev/null +++ b/model-00025-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f562bb4cc3571b9196b284603cfece1fdff2627f791758577c65af624096467d +size 503325464 diff --git a/model-00026-of-00081.safetensors b/model-00026-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..07da6692dcd0d6d8a857b24a34707f918ba2a138 --- /dev/null +++ b/model-00026-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1de71e773c33d5e1f7266a310e6abc2423b5ccbb708bc4a9769f1ec5f74883a6 +size 436225120 diff --git a/model-00027-of-00081.safetensors b/model-00027-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0954ed7f982324eb9d444ebd362182d1e9534a1c --- /dev/null +++ b/model-00027-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4ee223b62ec2ac18736e9159be39dd5b13ac3781f1367b28a79508022408e110 +size 469762600 diff --git a/model-00028-of-00081.safetensors b/model-00028-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..62b4399227472b307db5c715aa5a88f7205ce97d --- /dev/null +++ b/model-00028-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:976683d6e62ab38b1c78c492e374389c1c3d433518d6a740ed145860efdcd7e5 +size 436225120 diff --git a/model-00029-of-00081.safetensors b/model-00029-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..df6f0571f211f2bab9f68f2960ebbdc31e5abe7a --- /dev/null +++ b/model-00029-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ed0310dfee9a37e45ec23473f1abf8742d399ed946ee2fa36d14aadbe5cd26b7 +size 520102904 diff --git a/model-00030-of-00081.safetensors b/model-00030-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..36044556931c32945d4bd164a1d49271ad50a9c5 --- /dev/null +++ b/model-00030-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c1dbfa727560e175ab9abcf0c9b85d0b8a341c6f70fad5a889ec4a0090b49444 +size 503325464 diff --git a/model-00031-of-00081.safetensors b/model-00031-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b44418d0ce2db230e5c424c975b8e69e1b61caff --- /dev/null +++ b/model-00031-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef4f4801312e56a5e2ba4753f1c1f72fe2ec99db3f34a0022b7d1a26304b1f8a +size 436225120 diff --git a/model-00032-of-00081.safetensors b/model-00032-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..601b8b07fe9292af0ebd525813595dfb27e46749 --- /dev/null +++ b/model-00032-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b3e470a8b7c865c50c844e0252a6007ff1afc5c5dfaed19d2de330f483ac7d85 +size 469762600 diff --git a/model-00033-of-00081.safetensors b/model-00033-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..167b92cab9ce87466de0eaa09cada802ca50cb5c --- /dev/null +++ b/model-00033-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77cf16ef12160b8cdb19508cdff3bb247bbbbceb4c50957d63f198cc20940e20 +size 436225120 diff --git a/model-00034-of-00081.safetensors b/model-00034-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..724d6829469773f0cdaa83de4b1ad239803bd0d1 --- /dev/null +++ b/model-00034-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e713ce5b456dc52358f10a71040e632f563f4af6178bf82771788a977a0b9ad1 +size 520102904 diff --git a/model-00035-of-00081.safetensors b/model-00035-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c286f19c4cb80c4767ae5e878d5ae2de9e080753 --- /dev/null +++ b/model-00035-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c40622d04fd28a2bb1d4260a27ed9f9bfd3530e9b823b1e4b0d22ca419cd7f8 +size 503325464 diff --git a/model-00036-of-00081.safetensors b/model-00036-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4d0d1e01ef292deb963d5681196dc7ef36b33625 --- /dev/null +++ b/model-00036-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:01e5e725b38d0f42be7092c8f29fc226ed5a0ad3812a21f443f1cb9e8ea63a4d +size 436225120 diff --git a/model-00037-of-00081.safetensors b/model-00037-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..2771b7efe505cb1101480b9a1671869160f3fb52 --- /dev/null +++ b/model-00037-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:425a1535a59182633725e9d366f334d56f76240f33c21c9142d330b0a147843b +size 469762600 diff --git a/model-00038-of-00081.safetensors b/model-00038-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e212daf91868a6eecbb90141ecb373e39d07667e --- /dev/null +++ b/model-00038-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:132a92c4454438ff1ace820ce10aa0dcbdc80ee09e95445eeab505bdb1d13420 +size 436225120 diff --git a/model-00039-of-00081.safetensors b/model-00039-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e7fd6153c320a735fb790c7ae5d421610743d1b8 --- /dev/null +++ b/model-00039-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b57e5b18a54cc71401c9728d57db3a8eefe3f57b16489b5a9e80bc65e5e8c049 +size 520102904 diff --git a/model-00040-of-00081.safetensors b/model-00040-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b1002e8738b9d965f9c63dd42720ebc850d67c36 --- /dev/null +++ b/model-00040-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4b1b3ec383b47f5701d35dfa66c38c340ce7ed38da1040b7cca8868783dda27e +size 503325464 diff --git a/model-00041-of-00081.safetensors b/model-00041-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c55ef2524632560a6143acd617b72b3e3c922ecf --- /dev/null +++ b/model-00041-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a347fc3e0ce3ee8f01df3da9d7d3de0a0ad18118214745ba34f8fae49cb9a2fe +size 436225120 diff --git a/model-00042-of-00081.safetensors b/model-00042-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..bcb8b60a1127cbd9721c02611c5aa0172b3e4193 --- /dev/null +++ b/model-00042-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:48851f0b7f3224aaed3e8c24e6a87c0e705da4c269ac73ec696b15dc61d15291 +size 469762600 diff --git a/model-00043-of-00081.safetensors b/model-00043-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ccca91850ff999ab7020a6ac87528b702f2dd73d --- /dev/null +++ b/model-00043-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0ef0c1bd4936e59cdc8e375c64eec823b92a37004cb31aaa1691b57c2aaf7edf +size 436225120 diff --git a/model-00044-of-00081.safetensors b/model-00044-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..495e8afae9323f2a15eeb3634ac884d9c68fc595 --- /dev/null +++ b/model-00044-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ada1db1def9a795931fc054cffbfd2549053a86232895cd34be7ccec36789898 +size 520102904 diff --git a/model-00045-of-00081.safetensors b/model-00045-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..904582908ea5ff160128650fdb4b6d82605148fe --- /dev/null +++ b/model-00045-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64bcab2ff40469833ff26eb20c32718204d11022c4fade138ff946c5123b189f +size 503325464 diff --git a/model-00046-of-00081.safetensors b/model-00046-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c3cc752475af21456455fa055a26bd279ffd095d --- /dev/null +++ b/model-00046-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d4726aef4417630708e9019dbff0eb9bcdfa4d118f28ad8e92115f152e099c49 +size 436225120 diff --git a/model-00047-of-00081.safetensors b/model-00047-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a06fb62efa9a1812fa1017a119ddf79c5a17ec5a --- /dev/null +++ b/model-00047-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e4f2be5c276a7404363048239b748769ea04fe0827f40046d2144e2e70bdc34f +size 469762600 diff --git a/model-00048-of-00081.safetensors b/model-00048-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d994f61c0bf82f202b380e335d2fea449e2f3d8f --- /dev/null +++ b/model-00048-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dbbe7c3a3b7e9a4768f99ebd435187c333c689adf7e0ad6a8d05ba7eb373af9e +size 436225120 diff --git a/model-00049-of-00081.safetensors b/model-00049-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0bed3cf3e8eb103d08474e12afc83004746e9c3f --- /dev/null +++ b/model-00049-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a181fcfdf181acde90d5af6ad2e22340b0a1f9023b8e4af31404c9acda3293e0 +size 520102904 diff --git a/model-00050-of-00081.safetensors b/model-00050-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e572e36081335b3dbfe3edfecb9ab23d99fec56b --- /dev/null +++ b/model-00050-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b3ffabf5de26001887b3306b036f7e5e39174714d6f26871e88756b0a49f7ccc +size 503325464 diff --git a/model-00051-of-00081.safetensors b/model-00051-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..715e8793e3521c829c248252420f11ea27f9c768 --- /dev/null +++ b/model-00051-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f5955d55a6366cd2dfed9adf6ce52eca3c7d54c634445a902a006ed4a1f8778e +size 436225120 diff --git a/model-00052-of-00081.safetensors b/model-00052-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4067fb63bb38ce6746abc8b2b7427508d18a021e --- /dev/null +++ b/model-00052-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc7a6d691402a65d84ca31026795516c089560515fbbb8280ed94cee0022b7d5 +size 469762600 diff --git a/model-00053-of-00081.safetensors b/model-00053-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d1f587f874f8d35c9fa47a19ba355a59a9d8feb9 --- /dev/null +++ b/model-00053-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c997cc02e921ad13d833517d3d2d109d9cdacf5c76f5ac885c7df41f9edb4d7c +size 436225120 diff --git a/model-00054-of-00081.safetensors b/model-00054-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e92af8aaab29cc3cd121836fd7b529cb1407d77a --- /dev/null +++ b/model-00054-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:606a645f3df65c3fcabaabf40a2438146d7df3fef641c3362314255cd8e64c56 +size 520102904 diff --git a/model-00055-of-00081.safetensors b/model-00055-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..9a2f5dbd8258cd0108d38594eb4f69e73222cbc8 --- /dev/null +++ b/model-00055-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f91bdb69316edcb24bdecea18df4092c1bc9f868a840a0c0308df4d2b0a73d62 +size 503325464 diff --git a/model-00056-of-00081.safetensors b/model-00056-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7eadb0642b0ca35d3d0fea784eea6b4c1590fec3 --- /dev/null +++ b/model-00056-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7eaa60029ef954807bfe607e967defad241b49de558800a6f41220687a9d32dc +size 436225120 diff --git a/model-00057-of-00081.safetensors b/model-00057-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c60a9932e33f209c1151d3952941989e99c8418a --- /dev/null +++ b/model-00057-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a9be273953ccbdca8870cbfaa1d4d9e5683d7f1411876e600c83b0da80916808 +size 469762600 diff --git a/model-00058-of-00081.safetensors b/model-00058-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..daf65a6383d706e0c7cbfa2eb0e69a34c237cda4 --- /dev/null +++ b/model-00058-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:036b13e38def49abc7ad35eb48b244a5b08faf3230fbdba295b1360cb7e3f888 +size 436225120 diff --git a/model-00059-of-00081.safetensors b/model-00059-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c4bad712d302d8355626e26803c13b4818f3619e --- /dev/null +++ b/model-00059-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b6242b81d9e3cb858b508d7ef31baf1e94cc51e07ac0efd8a775da26f2bbe58 +size 520102904 diff --git a/model-00060-of-00081.safetensors b/model-00060-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..229dc4fe1439f7567a7af281053b2835e8ec6456 --- /dev/null +++ b/model-00060-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2e4d896375f96927b156dd528c921d09c4197d550173348ec0a368e6c4f98158 +size 503325464 diff --git a/model-00061-of-00081.safetensors b/model-00061-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..fbfee4dddf4fdd705564c105ffeb7e74793e1d17 --- /dev/null +++ b/model-00061-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6179e8e882f5dda0718902114da2835efc34fe9d76174b0ca5a83e94f16000ad +size 436225120 diff --git a/model-00062-of-00081.safetensors b/model-00062-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..361c2cdb1ea093018368e16503f8d4611e1ba593 --- /dev/null +++ b/model-00062-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4a3fa56db7b18a3bec451cfe9219ae52bf29f8fa975ddd0656c86f19fab75756 +size 469762600 diff --git a/model-00063-of-00081.safetensors b/model-00063-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..676b09452b191991be24e86f855ff05762bd242b --- /dev/null +++ b/model-00063-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:805d4ca22fa5ed6feca7bba9c7543ae7b9de4e8fa913c293f33cd608e5b08813 +size 436225120 diff --git a/model-00064-of-00081.safetensors b/model-00064-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8a8cb7d7770027ae1fda45af7d0c2f3c8eb99bfc --- /dev/null +++ b/model-00064-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:99c2e97bd960aa79804aa3453b8a91328df2c3d31247d7f34dbd85d1179ac6ff +size 520102904 diff --git a/model-00065-of-00081.safetensors b/model-00065-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..52ab5b6ee11898f84036aab09e4de869ff478523 --- /dev/null +++ b/model-00065-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:87517f79f49b63ff45d1d494061c730afe8c5df491ebb4a4af57eadc3cdb4601 +size 503325464 diff --git a/model-00066-of-00081.safetensors b/model-00066-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b2aa112393aa4f3a3b0aa855a99ad04dacb80d86 --- /dev/null +++ b/model-00066-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:138cdb33817f164b591e49a02f6176e7444593ecc4c5efbc80d2923d01b91af8 +size 436225120 diff --git a/model-00067-of-00081.safetensors b/model-00067-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f793e31e6ccb0dc2d31a66bf30c5f37c641cf93f --- /dev/null +++ b/model-00067-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:193a384b909982c82f509dd6310a33653ef088a021d3829774239a4150964717 +size 469762600 diff --git a/model-00068-of-00081.safetensors b/model-00068-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..be927ae5e89d6817cf6c0cd2d0cf87d29c352a74 --- /dev/null +++ b/model-00068-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c77678f67c902d8e91154033ce3352aa2ed102fbd49d69d99c289d8c799717af +size 436225120 diff --git a/model-00069-of-00081.safetensors b/model-00069-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..22f62d02e7722ec8e60d67c1344bc38958189259 --- /dev/null +++ b/model-00069-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:376e8e230c01efc80a0d7489299351cec1eaa62006455b445ff7ed290d419485 +size 520102904 diff --git a/model-00070-of-00081.safetensors b/model-00070-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..43e5b0478e108f2368c5ccdcbcf1287072cae993 --- /dev/null +++ b/model-00070-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ab730bdfc55b7c48e8a47a522c87addf0d0de0aa5a4c1d6e4eee3aa81477f36 +size 503325464 diff --git a/model-00071-of-00081.safetensors b/model-00071-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4929f01ea079746f77534dd4342e7e6114e3d781 --- /dev/null +++ b/model-00071-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8150cb9cb950ab29e7218bdf409ca3f7a7e897852518f1963ac185a9d5857ba9 +size 436225120 diff --git a/model-00072-of-00081.safetensors b/model-00072-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..6b038f3b62a865dd0e4004dfde432c3398279844 --- /dev/null +++ b/model-00072-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8dd94315ef9b66d1458f49ea9af2300070f370062f371ec4c57650b6101b00cc +size 469762600 diff --git a/model-00073-of-00081.safetensors b/model-00073-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..481c6c31af6e2fbc8180ddc81207ebf7930ab3cb --- /dev/null +++ b/model-00073-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f4fe7e6d337856812839c6805615a71a67b05953f53317dfce195c27f0050d45 +size 436225120 diff --git a/model-00074-of-00081.safetensors b/model-00074-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..21586d11f6ce3d503304641c14b69ca7f48493d7 --- /dev/null +++ b/model-00074-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b4e77377e5028744f38ab99a086ecf5cf28f04d9e62c92c44b1397011b123677 +size 520102904 diff --git a/model-00075-of-00081.safetensors b/model-00075-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..df3c73ced8f43c9c1bc11f80af7ba1edda2d0779 --- /dev/null +++ b/model-00075-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0bbb1344306a633909e9b0b35a04af33e9ff7ec62ae99bcafdaff94c34e5fd24 +size 503325464 diff --git a/model-00076-of-00081.safetensors b/model-00076-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1ad1b6fb47dabc8ae4ec861c5c370dc714752b18 --- /dev/null +++ b/model-00076-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e5146274adba9673afb5ec99c8ffc66739912f0f14f4045ecba22a07d055a14b +size 436225120 diff --git a/model-00077-of-00081.safetensors b/model-00077-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c254e6d8e1413ff728eb09707e98f2179a249294 --- /dev/null +++ b/model-00077-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abfb17d13ad2c5b943d97b19c861ed4f7054a0ab44cc570660e5d74bd269d799 +size 469762600 diff --git a/model-00078-of-00081.safetensors b/model-00078-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..65faaf81b836f5cc9a3938a6bf1a4dd805e795bd --- /dev/null +++ b/model-00078-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8b1438a2b0bdf2b874f064d6eb80c1dcb9fb87293e6c1136b002701f74de9da1 +size 436225120 diff --git a/model-00079-of-00081.safetensors b/model-00079-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..63571b0030ebb81d6f6082ca8f33fe222b0d4991 --- /dev/null +++ b/model-00079-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce0eb8beb30dc63e2fd40c398001acbb46f9582d98d221cfdf45bf9fd502c1cc +size 520102904 diff --git a/model-00080-of-00081.safetensors b/model-00080-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..908a47b86a1eb964c00f473ac273311ad924483e --- /dev/null +++ b/model-00080-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2e8490f724f48ca5e4b002314d9ea45e90e5473a1c8ff79c97cff8f3de42bb9d +size 503325464 diff --git a/model-00081-of-00081.safetensors b/model-00081-of-00081.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..88f4b9f84fb395190cbaab777194bcefe53a9a18 --- /dev/null +++ b/model-00081-of-00081.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a4fe66e8f9a3c9f008cecaf23bf83a2f5cf32d278cd91c74f9714d5f834fbc06 +size 4530943800 diff --git a/model.safetensors.index.json b/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..738510e260b0a2c981d93d4a12c5f14c4bacf3f2 --- /dev/null +++ b/model.safetensors.index.json @@ -0,0 +1,635 @@ +{ + "metadata": { + "total_size": 38375268352, + "mergekit_version": "0.1.4" + }, + "weight_map": { + "model.embed_tokens.weight": "model-00001-of-00081.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00081.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00081.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00081.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00081.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00081.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00081.safetensors", + "model.layers.0.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00081.safetensors", + "model.layers.0.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00081.safetensors", + "model.layers.0.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00081.safetensors", + "model.layers.0.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00081.safetensors", + "model.layers.0.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00081.safetensors", + "model.layers.0.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00081.safetensors", + "model.layers.1.input_layernorm.weight": "model-00003-of-00081.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00003-of-00081.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00003-of-00081.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00003-of-00081.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00003-of-00081.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00003-of-00081.safetensors", + "model.layers.1.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00081.safetensors", + "model.layers.1.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00081.safetensors", + "model.layers.1.block_sparse_moe.experts.0.w1.weight": "model-00004-of-00081.safetensors", + "model.layers.1.block_sparse_moe.experts.1.w1.weight": "model-00004-of-00081.safetensors", + "model.layers.1.block_sparse_moe.experts.0.w2.weight": "model-00004-of-00081.safetensors", + "model.layers.1.block_sparse_moe.experts.1.w2.weight": "model-00004-of-00081.safetensors", + "model.layers.2.input_layernorm.weight": "model-00004-of-00081.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00004-of-00081.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00004-of-00081.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00004-of-00081.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00005-of-00081.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00005-of-00081.safetensors", + "model.layers.2.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00081.safetensors", + "model.layers.2.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00081.safetensors", + "model.layers.2.block_sparse_moe.experts.0.w1.weight": "model-00005-of-00081.safetensors", + "model.layers.2.block_sparse_moe.experts.1.w1.weight": "model-00005-of-00081.safetensors", + "model.layers.2.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00081.safetensors", + "model.layers.2.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00081.safetensors", + "model.layers.3.input_layernorm.weight": "model-00006-of-00081.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00006-of-00081.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00006-of-00081.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00006-of-00081.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00006-of-00081.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00006-of-00081.safetensors", + "model.layers.3.block_sparse_moe.experts.0.w3.weight": "model-00006-of-00081.safetensors", + "model.layers.3.block_sparse_moe.experts.1.w3.weight": "model-00007-of-00081.safetensors", + "model.layers.3.block_sparse_moe.experts.0.w1.weight": "model-00007-of-00081.safetensors", + "model.layers.3.block_sparse_moe.experts.1.w1.weight": "model-00007-of-00081.safetensors", + "model.layers.3.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00081.safetensors", + "model.layers.3.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00081.safetensors", + "model.layers.4.input_layernorm.weight": "model-00008-of-00081.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00008-of-00081.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00008-of-00081.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00008-of-00081.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00008-of-00081.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00008-of-00081.safetensors", + "model.layers.4.block_sparse_moe.experts.0.w3.weight": "model-00008-of-00081.safetensors", + "model.layers.4.block_sparse_moe.experts.1.w3.weight": "model-00008-of-00081.safetensors", + "model.layers.4.block_sparse_moe.experts.0.w1.weight": "model-00009-of-00081.safetensors", + "model.layers.4.block_sparse_moe.experts.1.w1.weight": "model-00009-of-00081.safetensors", + "model.layers.4.block_sparse_moe.experts.0.w2.weight": "model-00009-of-00081.safetensors", + "model.layers.4.block_sparse_moe.experts.1.w2.weight": "model-00009-of-00081.safetensors", + "model.layers.5.input_layernorm.weight": "model-00009-of-00081.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00009-of-00081.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00009-of-00081.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00009-of-00081.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00010-of-00081.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00010-of-00081.safetensors", + "model.layers.5.block_sparse_moe.experts.0.w3.weight": "model-00010-of-00081.safetensors", + "model.layers.5.block_sparse_moe.experts.1.w3.weight": "model-00010-of-00081.safetensors", + "model.layers.5.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00081.safetensors", + "model.layers.5.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00081.safetensors", + "model.layers.5.block_sparse_moe.experts.0.w2.weight": "model-00011-of-00081.safetensors", + "model.layers.5.block_sparse_moe.experts.1.w2.weight": "model-00011-of-00081.safetensors", + "model.layers.6.input_layernorm.weight": "model-00011-of-00081.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00011-of-00081.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00011-of-00081.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00011-of-00081.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00011-of-00081.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00011-of-00081.safetensors", + "model.layers.6.block_sparse_moe.experts.0.w3.weight": "model-00011-of-00081.safetensors", + "model.layers.6.block_sparse_moe.experts.1.w3.weight": "model-00012-of-00081.safetensors", + "model.layers.6.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00081.safetensors", + "model.layers.6.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00081.safetensors", + "model.layers.6.block_sparse_moe.experts.0.w2.weight": "model-00012-of-00081.safetensors", + "model.layers.6.block_sparse_moe.experts.1.w2.weight": "model-00013-of-00081.safetensors", + "model.layers.7.input_layernorm.weight": "model-00013-of-00081.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00013-of-00081.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00013-of-00081.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00013-of-00081.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00013-of-00081.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00013-of-00081.safetensors", + "model.layers.7.block_sparse_moe.experts.0.w3.weight": "model-00013-of-00081.safetensors", + "model.layers.7.block_sparse_moe.experts.1.w3.weight": "model-00013-of-00081.safetensors", + "model.layers.7.block_sparse_moe.experts.0.w1.weight": "model-00014-of-00081.safetensors", + "model.layers.7.block_sparse_moe.experts.1.w1.weight": "model-00014-of-00081.safetensors", + "model.layers.7.block_sparse_moe.experts.0.w2.weight": "model-00014-of-00081.safetensors", + "model.layers.7.block_sparse_moe.experts.1.w2.weight": "model-00014-of-00081.safetensors", + "model.layers.8.input_layernorm.weight": "model-00014-of-00081.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00014-of-00081.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00014-of-00081.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00014-of-00081.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00015-of-00081.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00015-of-00081.safetensors", + "model.layers.8.block_sparse_moe.experts.0.w3.weight": "model-00015-of-00081.safetensors", + "model.layers.8.block_sparse_moe.experts.1.w3.weight": "model-00015-of-00081.safetensors", + "model.layers.8.block_sparse_moe.experts.0.w1.weight": "model-00015-of-00081.safetensors", + "model.layers.8.block_sparse_moe.experts.1.w1.weight": "model-00015-of-00081.safetensors", + "model.layers.8.block_sparse_moe.experts.0.w2.weight": "model-00016-of-00081.safetensors", + "model.layers.8.block_sparse_moe.experts.1.w2.weight": "model-00016-of-00081.safetensors", + "model.layers.9.input_layernorm.weight": "model-00016-of-00081.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00016-of-00081.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00016-of-00081.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00016-of-00081.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00016-of-00081.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00016-of-00081.safetensors", + "model.layers.9.block_sparse_moe.experts.0.w3.weight": "model-00016-of-00081.safetensors", + "model.layers.9.block_sparse_moe.experts.1.w3.weight": "model-00017-of-00081.safetensors", + "model.layers.9.block_sparse_moe.experts.0.w1.weight": "model-00017-of-00081.safetensors", + "model.layers.9.block_sparse_moe.experts.1.w1.weight": "model-00017-of-00081.safetensors", + "model.layers.9.block_sparse_moe.experts.0.w2.weight": "model-00017-of-00081.safetensors", + "model.layers.9.block_sparse_moe.experts.1.w2.weight": "model-00018-of-00081.safetensors", + "model.layers.10.input_layernorm.weight": "model-00018-of-00081.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00018-of-00081.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00018-of-00081.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00018-of-00081.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00018-of-00081.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00018-of-00081.safetensors", + "model.layers.10.block_sparse_moe.experts.0.w3.weight": "model-00018-of-00081.safetensors", + "model.layers.10.block_sparse_moe.experts.1.w3.weight": "model-00018-of-00081.safetensors", + "model.layers.10.block_sparse_moe.experts.0.w1.weight": "model-00019-of-00081.safetensors", + "model.layers.10.block_sparse_moe.experts.1.w1.weight": "model-00019-of-00081.safetensors", + "model.layers.10.block_sparse_moe.experts.0.w2.weight": "model-00019-of-00081.safetensors", + "model.layers.10.block_sparse_moe.experts.1.w2.weight": "model-00019-of-00081.safetensors", + "model.layers.11.input_layernorm.weight": "model-00019-of-00081.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00019-of-00081.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00019-of-00081.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00019-of-00081.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00020-of-00081.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00020-of-00081.safetensors", + "model.layers.11.block_sparse_moe.experts.0.w3.weight": "model-00020-of-00081.safetensors", + "model.layers.11.block_sparse_moe.experts.1.w3.weight": "model-00020-of-00081.safetensors", + "model.layers.11.block_sparse_moe.experts.0.w1.weight": "model-00020-of-00081.safetensors", + "model.layers.11.block_sparse_moe.experts.1.w1.weight": "model-00020-of-00081.safetensors", + "model.layers.11.block_sparse_moe.experts.0.w2.weight": "model-00021-of-00081.safetensors", + "model.layers.11.block_sparse_moe.experts.1.w2.weight": "model-00021-of-00081.safetensors", + "model.layers.12.input_layernorm.weight": "model-00021-of-00081.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00021-of-00081.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00021-of-00081.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00021-of-00081.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00021-of-00081.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00021-of-00081.safetensors", + "model.layers.12.block_sparse_moe.experts.0.w3.weight": "model-00021-of-00081.safetensors", + "model.layers.12.block_sparse_moe.experts.1.w3.weight": "model-00022-of-00081.safetensors", + "model.layers.12.block_sparse_moe.experts.0.w1.weight": "model-00022-of-00081.safetensors", + "model.layers.12.block_sparse_moe.experts.1.w1.weight": "model-00022-of-00081.safetensors", + "model.layers.12.block_sparse_moe.experts.0.w2.weight": "model-00022-of-00081.safetensors", + "model.layers.12.block_sparse_moe.experts.1.w2.weight": "model-00023-of-00081.safetensors", + "model.layers.13.input_layernorm.weight": "model-00023-of-00081.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00023-of-00081.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00023-of-00081.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00023-of-00081.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00023-of-00081.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00023-of-00081.safetensors", + "model.layers.13.block_sparse_moe.experts.0.w3.weight": "model-00023-of-00081.safetensors", + "model.layers.13.block_sparse_moe.experts.1.w3.weight": "model-00023-of-00081.safetensors", + "model.layers.13.block_sparse_moe.experts.0.w1.weight": "model-00024-of-00081.safetensors", + "model.layers.13.block_sparse_moe.experts.1.w1.weight": "model-00024-of-00081.safetensors", + "model.layers.13.block_sparse_moe.experts.0.w2.weight": "model-00024-of-00081.safetensors", + "model.layers.13.block_sparse_moe.experts.1.w2.weight": "model-00024-of-00081.safetensors", + "model.layers.14.input_layernorm.weight": "model-00024-of-00081.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00024-of-00081.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00024-of-00081.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00024-of-00081.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00025-of-00081.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00025-of-00081.safetensors", + "model.layers.14.block_sparse_moe.experts.0.w3.weight": "model-00025-of-00081.safetensors", + "model.layers.14.block_sparse_moe.experts.1.w3.weight": "model-00025-of-00081.safetensors", + "model.layers.14.block_sparse_moe.experts.0.w1.weight": "model-00025-of-00081.safetensors", + "model.layers.14.block_sparse_moe.experts.1.w1.weight": "model-00025-of-00081.safetensors", + "model.layers.14.block_sparse_moe.experts.0.w2.weight": "model-00026-of-00081.safetensors", + "model.layers.14.block_sparse_moe.experts.1.w2.weight": "model-00026-of-00081.safetensors", + "model.layers.15.input_layernorm.weight": "model-00026-of-00081.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00026-of-00081.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00026-of-00081.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00026-of-00081.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00026-of-00081.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00026-of-00081.safetensors", + "model.layers.15.block_sparse_moe.experts.0.w3.weight": "model-00026-of-00081.safetensors", + "model.layers.15.block_sparse_moe.experts.1.w3.weight": "model-00027-of-00081.safetensors", + "model.layers.15.block_sparse_moe.experts.0.w1.weight": "model-00027-of-00081.safetensors", + "model.layers.15.block_sparse_moe.experts.1.w1.weight": "model-00027-of-00081.safetensors", + "model.layers.15.block_sparse_moe.experts.0.w2.weight": "model-00027-of-00081.safetensors", + "model.layers.15.block_sparse_moe.experts.1.w2.weight": "model-00028-of-00081.safetensors", + "model.layers.16.input_layernorm.weight": "model-00028-of-00081.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00028-of-00081.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00028-of-00081.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00028-of-00081.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00028-of-00081.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00028-of-00081.safetensors", + "model.layers.16.block_sparse_moe.experts.0.w3.weight": "model-00028-of-00081.safetensors", + "model.layers.16.block_sparse_moe.experts.1.w3.weight": "model-00028-of-00081.safetensors", + "model.layers.16.block_sparse_moe.experts.0.w1.weight": "model-00029-of-00081.safetensors", + "model.layers.16.block_sparse_moe.experts.1.w1.weight": "model-00029-of-00081.safetensors", + "model.layers.16.block_sparse_moe.experts.0.w2.weight": "model-00029-of-00081.safetensors", + "model.layers.16.block_sparse_moe.experts.1.w2.weight": "model-00029-of-00081.safetensors", + "model.layers.17.input_layernorm.weight": "model-00029-of-00081.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00029-of-00081.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00029-of-00081.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00029-of-00081.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00030-of-00081.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00030-of-00081.safetensors", + "model.layers.17.block_sparse_moe.experts.0.w3.weight": "model-00030-of-00081.safetensors", + "model.layers.17.block_sparse_moe.experts.1.w3.weight": "model-00030-of-00081.safetensors", + "model.layers.17.block_sparse_moe.experts.0.w1.weight": "model-00030-of-00081.safetensors", + "model.layers.17.block_sparse_moe.experts.1.w1.weight": "model-00030-of-00081.safetensors", + "model.layers.17.block_sparse_moe.experts.0.w2.weight": "model-00031-of-00081.safetensors", + "model.layers.17.block_sparse_moe.experts.1.w2.weight": "model-00031-of-00081.safetensors", + "model.layers.18.input_layernorm.weight": "model-00031-of-00081.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00031-of-00081.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00031-of-00081.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00031-of-00081.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00031-of-00081.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00031-of-00081.safetensors", + "model.layers.18.block_sparse_moe.experts.0.w3.weight": "model-00031-of-00081.safetensors", + "model.layers.18.block_sparse_moe.experts.1.w3.weight": "model-00032-of-00081.safetensors", + "model.layers.18.block_sparse_moe.experts.0.w1.weight": "model-00032-of-00081.safetensors", + "model.layers.18.block_sparse_moe.experts.1.w1.weight": "model-00032-of-00081.safetensors", + "model.layers.18.block_sparse_moe.experts.0.w2.weight": "model-00032-of-00081.safetensors", + "model.layers.18.block_sparse_moe.experts.1.w2.weight": "model-00033-of-00081.safetensors", + "model.layers.19.input_layernorm.weight": "model-00033-of-00081.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00033-of-00081.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00033-of-00081.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00033-of-00081.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00033-of-00081.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00033-of-00081.safetensors", + "model.layers.19.block_sparse_moe.experts.0.w3.weight": "model-00033-of-00081.safetensors", + "model.layers.19.block_sparse_moe.experts.1.w3.weight": "model-00033-of-00081.safetensors", + "model.layers.19.block_sparse_moe.experts.0.w1.weight": "model-00034-of-00081.safetensors", + "model.layers.19.block_sparse_moe.experts.1.w1.weight": "model-00034-of-00081.safetensors", + "model.layers.19.block_sparse_moe.experts.0.w2.weight": "model-00034-of-00081.safetensors", + "model.layers.19.block_sparse_moe.experts.1.w2.weight": "model-00034-of-00081.safetensors", + "model.layers.20.input_layernorm.weight": "model-00034-of-00081.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00034-of-00081.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00034-of-00081.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00034-of-00081.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00035-of-00081.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00035-of-00081.safetensors", + "model.layers.20.block_sparse_moe.experts.0.w3.weight": "model-00035-of-00081.safetensors", + "model.layers.20.block_sparse_moe.experts.1.w3.weight": "model-00035-of-00081.safetensors", + "model.layers.20.block_sparse_moe.experts.0.w1.weight": "model-00035-of-00081.safetensors", + "model.layers.20.block_sparse_moe.experts.1.w1.weight": "model-00035-of-00081.safetensors", + "model.layers.20.block_sparse_moe.experts.0.w2.weight": "model-00036-of-00081.safetensors", + "model.layers.20.block_sparse_moe.experts.1.w2.weight": "model-00036-of-00081.safetensors", + "model.layers.21.input_layernorm.weight": "model-00036-of-00081.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00036-of-00081.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00036-of-00081.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00036-of-00081.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00036-of-00081.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00036-of-00081.safetensors", + "model.layers.21.block_sparse_moe.experts.0.w3.weight": "model-00036-of-00081.safetensors", + "model.layers.21.block_sparse_moe.experts.1.w3.weight": "model-00037-of-00081.safetensors", + "model.layers.21.block_sparse_moe.experts.0.w1.weight": "model-00037-of-00081.safetensors", + "model.layers.21.block_sparse_moe.experts.1.w1.weight": "model-00037-of-00081.safetensors", + "model.layers.21.block_sparse_moe.experts.0.w2.weight": "model-00037-of-00081.safetensors", + "model.layers.21.block_sparse_moe.experts.1.w2.weight": "model-00038-of-00081.safetensors", + "model.layers.22.input_layernorm.weight": "model-00038-of-00081.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00038-of-00081.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00038-of-00081.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00038-of-00081.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00038-of-00081.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00038-of-00081.safetensors", + "model.layers.22.block_sparse_moe.experts.0.w3.weight": "model-00038-of-00081.safetensors", + "model.layers.22.block_sparse_moe.experts.1.w3.weight": "model-00038-of-00081.safetensors", + "model.layers.22.block_sparse_moe.experts.0.w1.weight": "model-00039-of-00081.safetensors", + "model.layers.22.block_sparse_moe.experts.1.w1.weight": "model-00039-of-00081.safetensors", + "model.layers.22.block_sparse_moe.experts.0.w2.weight": "model-00039-of-00081.safetensors", + "model.layers.22.block_sparse_moe.experts.1.w2.weight": "model-00039-of-00081.safetensors", + "model.layers.23.input_layernorm.weight": "model-00039-of-00081.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00039-of-00081.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00039-of-00081.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00039-of-00081.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00040-of-00081.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00040-of-00081.safetensors", + "model.layers.23.block_sparse_moe.experts.0.w3.weight": "model-00040-of-00081.safetensors", + "model.layers.23.block_sparse_moe.experts.1.w3.weight": "model-00040-of-00081.safetensors", + "model.layers.23.block_sparse_moe.experts.0.w1.weight": "model-00040-of-00081.safetensors", + "model.layers.23.block_sparse_moe.experts.1.w1.weight": "model-00040-of-00081.safetensors", + "model.layers.23.block_sparse_moe.experts.0.w2.weight": "model-00041-of-00081.safetensors", + "model.layers.23.block_sparse_moe.experts.1.w2.weight": "model-00041-of-00081.safetensors", + "model.layers.24.input_layernorm.weight": "model-00041-of-00081.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00041-of-00081.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00041-of-00081.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00041-of-00081.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00041-of-00081.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00041-of-00081.safetensors", + "model.layers.24.block_sparse_moe.experts.0.w3.weight": "model-00041-of-00081.safetensors", + "model.layers.24.block_sparse_moe.experts.1.w3.weight": "model-00042-of-00081.safetensors", + "model.layers.24.block_sparse_moe.experts.0.w1.weight": "model-00042-of-00081.safetensors", + "model.layers.24.block_sparse_moe.experts.1.w1.weight": "model-00042-of-00081.safetensors", + "model.layers.24.block_sparse_moe.experts.0.w2.weight": "model-00042-of-00081.safetensors", + "model.layers.24.block_sparse_moe.experts.1.w2.weight": "model-00043-of-00081.safetensors", + "model.layers.25.input_layernorm.weight": "model-00043-of-00081.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00043-of-00081.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00043-of-00081.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00043-of-00081.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00043-of-00081.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00043-of-00081.safetensors", + "model.layers.25.block_sparse_moe.experts.0.w3.weight": "model-00043-of-00081.safetensors", + "model.layers.25.block_sparse_moe.experts.1.w3.weight": "model-00043-of-00081.safetensors", + "model.layers.25.block_sparse_moe.experts.0.w1.weight": "model-00044-of-00081.safetensors", + "model.layers.25.block_sparse_moe.experts.1.w1.weight": "model-00044-of-00081.safetensors", + "model.layers.25.block_sparse_moe.experts.0.w2.weight": "model-00044-of-00081.safetensors", + "model.layers.25.block_sparse_moe.experts.1.w2.weight": "model-00044-of-00081.safetensors", + "model.layers.26.input_layernorm.weight": "model-00044-of-00081.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00044-of-00081.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00044-of-00081.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00044-of-00081.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00045-of-00081.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00045-of-00081.safetensors", + "model.layers.26.block_sparse_moe.experts.0.w3.weight": "model-00045-of-00081.safetensors", + "model.layers.26.block_sparse_moe.experts.1.w3.weight": "model-00045-of-00081.safetensors", + "model.layers.26.block_sparse_moe.experts.0.w1.weight": "model-00045-of-00081.safetensors", + "model.layers.26.block_sparse_moe.experts.1.w1.weight": "model-00045-of-00081.safetensors", + "model.layers.26.block_sparse_moe.experts.0.w2.weight": "model-00046-of-00081.safetensors", + "model.layers.26.block_sparse_moe.experts.1.w2.weight": "model-00046-of-00081.safetensors", + "model.layers.27.input_layernorm.weight": "model-00046-of-00081.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00046-of-00081.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00046-of-00081.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00046-of-00081.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00046-of-00081.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00046-of-00081.safetensors", + "model.layers.27.block_sparse_moe.experts.0.w3.weight": "model-00046-of-00081.safetensors", + "model.layers.27.block_sparse_moe.experts.1.w3.weight": "model-00047-of-00081.safetensors", + "model.layers.27.block_sparse_moe.experts.0.w1.weight": "model-00047-of-00081.safetensors", + "model.layers.27.block_sparse_moe.experts.1.w1.weight": "model-00047-of-00081.safetensors", + "model.layers.27.block_sparse_moe.experts.0.w2.weight": "model-00047-of-00081.safetensors", + "model.layers.27.block_sparse_moe.experts.1.w2.weight": "model-00048-of-00081.safetensors", + "model.layers.28.input_layernorm.weight": "model-00048-of-00081.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00048-of-00081.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00048-of-00081.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00048-of-00081.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00048-of-00081.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00048-of-00081.safetensors", + "model.layers.28.block_sparse_moe.experts.0.w3.weight": "model-00048-of-00081.safetensors", + "model.layers.28.block_sparse_moe.experts.1.w3.weight": "model-00048-of-00081.safetensors", + "model.layers.28.block_sparse_moe.experts.0.w1.weight": "model-00049-of-00081.safetensors", + "model.layers.28.block_sparse_moe.experts.1.w1.weight": "model-00049-of-00081.safetensors", + "model.layers.28.block_sparse_moe.experts.0.w2.weight": "model-00049-of-00081.safetensors", + "model.layers.28.block_sparse_moe.experts.1.w2.weight": "model-00049-of-00081.safetensors", + "model.layers.29.input_layernorm.weight": "model-00049-of-00081.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00049-of-00081.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00049-of-00081.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00049-of-00081.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00050-of-00081.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00050-of-00081.safetensors", + "model.layers.29.block_sparse_moe.experts.0.w3.weight": "model-00050-of-00081.safetensors", + "model.layers.29.block_sparse_moe.experts.1.w3.weight": "model-00050-of-00081.safetensors", + "model.layers.29.block_sparse_moe.experts.0.w1.weight": "model-00050-of-00081.safetensors", + "model.layers.29.block_sparse_moe.experts.1.w1.weight": "model-00050-of-00081.safetensors", + "model.layers.29.block_sparse_moe.experts.0.w2.weight": "model-00051-of-00081.safetensors", + "model.layers.29.block_sparse_moe.experts.1.w2.weight": "model-00051-of-00081.safetensors", + "model.layers.30.input_layernorm.weight": "model-00051-of-00081.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00051-of-00081.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00051-of-00081.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00051-of-00081.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00051-of-00081.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00051-of-00081.safetensors", + "model.layers.30.block_sparse_moe.experts.0.w3.weight": "model-00051-of-00081.safetensors", + "model.layers.30.block_sparse_moe.experts.1.w3.weight": "model-00052-of-00081.safetensors", + "model.layers.30.block_sparse_moe.experts.0.w1.weight": "model-00052-of-00081.safetensors", + "model.layers.30.block_sparse_moe.experts.1.w1.weight": "model-00052-of-00081.safetensors", + "model.layers.30.block_sparse_moe.experts.0.w2.weight": "model-00052-of-00081.safetensors", + "model.layers.30.block_sparse_moe.experts.1.w2.weight": "model-00053-of-00081.safetensors", + "model.layers.31.input_layernorm.weight": "model-00053-of-00081.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00053-of-00081.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00053-of-00081.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00053-of-00081.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00053-of-00081.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00053-of-00081.safetensors", + "model.layers.31.block_sparse_moe.experts.0.w3.weight": "model-00053-of-00081.safetensors", + "model.layers.31.block_sparse_moe.experts.1.w3.weight": "model-00053-of-00081.safetensors", + "model.layers.31.block_sparse_moe.experts.0.w1.weight": "model-00054-of-00081.safetensors", + "model.layers.31.block_sparse_moe.experts.1.w1.weight": "model-00054-of-00081.safetensors", + "model.layers.31.block_sparse_moe.experts.0.w2.weight": "model-00054-of-00081.safetensors", + "model.layers.31.block_sparse_moe.experts.1.w2.weight": "model-00054-of-00081.safetensors", + "model.layers.32.input_layernorm.weight": "model-00054-of-00081.safetensors", + "model.layers.32.self_attn.q_proj.weight": "model-00054-of-00081.safetensors", + "model.layers.32.self_attn.k_proj.weight": "model-00054-of-00081.safetensors", + "model.layers.32.self_attn.v_proj.weight": "model-00054-of-00081.safetensors", + "model.layers.32.self_attn.o_proj.weight": "model-00055-of-00081.safetensors", + "model.layers.32.post_attention_layernorm.weight": "model-00055-of-00081.safetensors", + "model.layers.32.block_sparse_moe.experts.0.w3.weight": "model-00055-of-00081.safetensors", + "model.layers.32.block_sparse_moe.experts.1.w3.weight": "model-00055-of-00081.safetensors", + "model.layers.32.block_sparse_moe.experts.0.w1.weight": "model-00055-of-00081.safetensors", + "model.layers.32.block_sparse_moe.experts.1.w1.weight": "model-00055-of-00081.safetensors", + "model.layers.32.block_sparse_moe.experts.0.w2.weight": "model-00056-of-00081.safetensors", + "model.layers.32.block_sparse_moe.experts.1.w2.weight": "model-00056-of-00081.safetensors", + "model.layers.33.input_layernorm.weight": "model-00056-of-00081.safetensors", + "model.layers.33.self_attn.q_proj.weight": "model-00056-of-00081.safetensors", + "model.layers.33.self_attn.k_proj.weight": "model-00056-of-00081.safetensors", + "model.layers.33.self_attn.v_proj.weight": "model-00056-of-00081.safetensors", + "model.layers.33.self_attn.o_proj.weight": "model-00056-of-00081.safetensors", + "model.layers.33.post_attention_layernorm.weight": "model-00056-of-00081.safetensors", + "model.layers.33.block_sparse_moe.experts.0.w3.weight": "model-00056-of-00081.safetensors", + "model.layers.33.block_sparse_moe.experts.1.w3.weight": "model-00057-of-00081.safetensors", + "model.layers.33.block_sparse_moe.experts.0.w1.weight": "model-00057-of-00081.safetensors", + "model.layers.33.block_sparse_moe.experts.1.w1.weight": "model-00057-of-00081.safetensors", + "model.layers.33.block_sparse_moe.experts.0.w2.weight": "model-00057-of-00081.safetensors", + "model.layers.33.block_sparse_moe.experts.1.w2.weight": "model-00058-of-00081.safetensors", + "model.layers.34.input_layernorm.weight": "model-00058-of-00081.safetensors", + "model.layers.34.self_attn.q_proj.weight": "model-00058-of-00081.safetensors", + "model.layers.34.self_attn.k_proj.weight": "model-00058-of-00081.safetensors", + "model.layers.34.self_attn.v_proj.weight": "model-00058-of-00081.safetensors", + "model.layers.34.self_attn.o_proj.weight": "model-00058-of-00081.safetensors", + "model.layers.34.post_attention_layernorm.weight": "model-00058-of-00081.safetensors", + "model.layers.34.block_sparse_moe.experts.0.w3.weight": "model-00058-of-00081.safetensors", + "model.layers.34.block_sparse_moe.experts.1.w3.weight": "model-00058-of-00081.safetensors", + "model.layers.34.block_sparse_moe.experts.0.w1.weight": "model-00059-of-00081.safetensors", + "model.layers.34.block_sparse_moe.experts.1.w1.weight": "model-00059-of-00081.safetensors", + "model.layers.34.block_sparse_moe.experts.0.w2.weight": "model-00059-of-00081.safetensors", + "model.layers.34.block_sparse_moe.experts.1.w2.weight": "model-00059-of-00081.safetensors", + "model.layers.35.input_layernorm.weight": "model-00059-of-00081.safetensors", + "model.layers.35.self_attn.q_proj.weight": "model-00059-of-00081.safetensors", + "model.layers.35.self_attn.k_proj.weight": "model-00059-of-00081.safetensors", + "model.layers.35.self_attn.v_proj.weight": "model-00059-of-00081.safetensors", + "model.layers.35.self_attn.o_proj.weight": "model-00060-of-00081.safetensors", + "model.layers.35.post_attention_layernorm.weight": "model-00060-of-00081.safetensors", + "model.layers.35.block_sparse_moe.experts.0.w3.weight": "model-00060-of-00081.safetensors", + "model.layers.35.block_sparse_moe.experts.1.w3.weight": "model-00060-of-00081.safetensors", + "model.layers.35.block_sparse_moe.experts.0.w1.weight": "model-00060-of-00081.safetensors", + "model.layers.35.block_sparse_moe.experts.1.w1.weight": "model-00060-of-00081.safetensors", + "model.layers.35.block_sparse_moe.experts.0.w2.weight": "model-00061-of-00081.safetensors", + "model.layers.35.block_sparse_moe.experts.1.w2.weight": "model-00061-of-00081.safetensors", + "model.layers.36.input_layernorm.weight": "model-00061-of-00081.safetensors", + "model.layers.36.self_attn.q_proj.weight": "model-00061-of-00081.safetensors", + "model.layers.36.self_attn.k_proj.weight": "model-00061-of-00081.safetensors", + "model.layers.36.self_attn.v_proj.weight": "model-00061-of-00081.safetensors", + "model.layers.36.self_attn.o_proj.weight": "model-00061-of-00081.safetensors", + "model.layers.36.post_attention_layernorm.weight": "model-00061-of-00081.safetensors", + "model.layers.36.block_sparse_moe.experts.0.w3.weight": "model-00061-of-00081.safetensors", + "model.layers.36.block_sparse_moe.experts.1.w3.weight": "model-00062-of-00081.safetensors", + "model.layers.36.block_sparse_moe.experts.0.w1.weight": "model-00062-of-00081.safetensors", + "model.layers.36.block_sparse_moe.experts.1.w1.weight": "model-00062-of-00081.safetensors", + "model.layers.36.block_sparse_moe.experts.0.w2.weight": "model-00062-of-00081.safetensors", + "model.layers.36.block_sparse_moe.experts.1.w2.weight": "model-00063-of-00081.safetensors", + "model.layers.37.input_layernorm.weight": "model-00063-of-00081.safetensors", + "model.layers.37.self_attn.q_proj.weight": "model-00063-of-00081.safetensors", + "model.layers.37.self_attn.k_proj.weight": "model-00063-of-00081.safetensors", + "model.layers.37.self_attn.v_proj.weight": "model-00063-of-00081.safetensors", + "model.layers.37.self_attn.o_proj.weight": "model-00063-of-00081.safetensors", + "model.layers.37.post_attention_layernorm.weight": "model-00063-of-00081.safetensors", + "model.layers.37.block_sparse_moe.experts.0.w3.weight": "model-00063-of-00081.safetensors", + "model.layers.37.block_sparse_moe.experts.1.w3.weight": "model-00063-of-00081.safetensors", + "model.layers.37.block_sparse_moe.experts.0.w1.weight": "model-00064-of-00081.safetensors", + "model.layers.37.block_sparse_moe.experts.1.w1.weight": "model-00064-of-00081.safetensors", + "model.layers.37.block_sparse_moe.experts.0.w2.weight": "model-00064-of-00081.safetensors", + "model.layers.37.block_sparse_moe.experts.1.w2.weight": "model-00064-of-00081.safetensors", + "model.layers.38.input_layernorm.weight": "model-00064-of-00081.safetensors", + "model.layers.38.self_attn.q_proj.weight": "model-00064-of-00081.safetensors", + "model.layers.38.self_attn.k_proj.weight": "model-00064-of-00081.safetensors", + "model.layers.38.self_attn.v_proj.weight": "model-00064-of-00081.safetensors", + "model.layers.38.self_attn.o_proj.weight": "model-00065-of-00081.safetensors", + "model.layers.38.post_attention_layernorm.weight": "model-00065-of-00081.safetensors", + "model.layers.38.block_sparse_moe.experts.0.w3.weight": "model-00065-of-00081.safetensors", + "model.layers.38.block_sparse_moe.experts.1.w3.weight": "model-00065-of-00081.safetensors", + "model.layers.38.block_sparse_moe.experts.0.w1.weight": "model-00065-of-00081.safetensors", + "model.layers.38.block_sparse_moe.experts.1.w1.weight": "model-00065-of-00081.safetensors", + "model.layers.38.block_sparse_moe.experts.0.w2.weight": "model-00066-of-00081.safetensors", + "model.layers.38.block_sparse_moe.experts.1.w2.weight": "model-00066-of-00081.safetensors", + "model.layers.39.input_layernorm.weight": "model-00066-of-00081.safetensors", + "model.layers.39.self_attn.q_proj.weight": "model-00066-of-00081.safetensors", + "model.layers.39.self_attn.k_proj.weight": "model-00066-of-00081.safetensors", + "model.layers.39.self_attn.v_proj.weight": "model-00066-of-00081.safetensors", + "model.layers.39.self_attn.o_proj.weight": "model-00066-of-00081.safetensors", + "model.layers.39.post_attention_layernorm.weight": "model-00066-of-00081.safetensors", + "model.layers.39.block_sparse_moe.experts.0.w3.weight": "model-00066-of-00081.safetensors", + "model.layers.39.block_sparse_moe.experts.1.w3.weight": "model-00067-of-00081.safetensors", + "model.layers.39.block_sparse_moe.experts.0.w1.weight": "model-00067-of-00081.safetensors", + "model.layers.39.block_sparse_moe.experts.1.w1.weight": "model-00067-of-00081.safetensors", + "model.layers.39.block_sparse_moe.experts.0.w2.weight": "model-00067-of-00081.safetensors", + "model.layers.39.block_sparse_moe.experts.1.w2.weight": "model-00068-of-00081.safetensors", + "model.layers.40.input_layernorm.weight": "model-00068-of-00081.safetensors", + "model.layers.40.self_attn.q_proj.weight": "model-00068-of-00081.safetensors", + "model.layers.40.self_attn.k_proj.weight": "model-00068-of-00081.safetensors", + "model.layers.40.self_attn.v_proj.weight": "model-00068-of-00081.safetensors", + "model.layers.40.self_attn.o_proj.weight": "model-00068-of-00081.safetensors", + "model.layers.40.post_attention_layernorm.weight": "model-00068-of-00081.safetensors", + "model.layers.40.block_sparse_moe.experts.0.w3.weight": "model-00068-of-00081.safetensors", + "model.layers.40.block_sparse_moe.experts.1.w3.weight": "model-00068-of-00081.safetensors", + "model.layers.40.block_sparse_moe.experts.0.w1.weight": "model-00069-of-00081.safetensors", + "model.layers.40.block_sparse_moe.experts.1.w1.weight": "model-00069-of-00081.safetensors", + "model.layers.40.block_sparse_moe.experts.0.w2.weight": "model-00069-of-00081.safetensors", + "model.layers.40.block_sparse_moe.experts.1.w2.weight": "model-00069-of-00081.safetensors", + "model.layers.41.input_layernorm.weight": "model-00069-of-00081.safetensors", + "model.layers.41.self_attn.q_proj.weight": "model-00069-of-00081.safetensors", + "model.layers.41.self_attn.k_proj.weight": "model-00069-of-00081.safetensors", + "model.layers.41.self_attn.v_proj.weight": "model-00069-of-00081.safetensors", + "model.layers.41.self_attn.o_proj.weight": "model-00070-of-00081.safetensors", + "model.layers.41.post_attention_layernorm.weight": "model-00070-of-00081.safetensors", + "model.layers.41.block_sparse_moe.experts.0.w3.weight": "model-00070-of-00081.safetensors", + "model.layers.41.block_sparse_moe.experts.1.w3.weight": "model-00070-of-00081.safetensors", + "model.layers.41.block_sparse_moe.experts.0.w1.weight": "model-00070-of-00081.safetensors", + "model.layers.41.block_sparse_moe.experts.1.w1.weight": "model-00070-of-00081.safetensors", + "model.layers.41.block_sparse_moe.experts.0.w2.weight": "model-00071-of-00081.safetensors", + "model.layers.41.block_sparse_moe.experts.1.w2.weight": "model-00071-of-00081.safetensors", + "model.layers.42.input_layernorm.weight": "model-00071-of-00081.safetensors", + "model.layers.42.self_attn.q_proj.weight": "model-00071-of-00081.safetensors", + "model.layers.42.self_attn.k_proj.weight": "model-00071-of-00081.safetensors", + "model.layers.42.self_attn.v_proj.weight": "model-00071-of-00081.safetensors", + "model.layers.42.self_attn.o_proj.weight": "model-00071-of-00081.safetensors", + "model.layers.42.post_attention_layernorm.weight": "model-00071-of-00081.safetensors", + "model.layers.42.block_sparse_moe.experts.0.w3.weight": "model-00071-of-00081.safetensors", + "model.layers.42.block_sparse_moe.experts.1.w3.weight": "model-00072-of-00081.safetensors", + "model.layers.42.block_sparse_moe.experts.0.w1.weight": "model-00072-of-00081.safetensors", + "model.layers.42.block_sparse_moe.experts.1.w1.weight": "model-00072-of-00081.safetensors", + "model.layers.42.block_sparse_moe.experts.0.w2.weight": "model-00072-of-00081.safetensors", + "model.layers.42.block_sparse_moe.experts.1.w2.weight": "model-00073-of-00081.safetensors", + "model.layers.43.input_layernorm.weight": "model-00073-of-00081.safetensors", + "model.layers.43.self_attn.q_proj.weight": "model-00073-of-00081.safetensors", + "model.layers.43.self_attn.k_proj.weight": "model-00073-of-00081.safetensors", + "model.layers.43.self_attn.v_proj.weight": "model-00073-of-00081.safetensors", + "model.layers.43.self_attn.o_proj.weight": "model-00073-of-00081.safetensors", + "model.layers.43.post_attention_layernorm.weight": "model-00073-of-00081.safetensors", + "model.layers.43.block_sparse_moe.experts.0.w3.weight": "model-00073-of-00081.safetensors", + "model.layers.43.block_sparse_moe.experts.1.w3.weight": "model-00073-of-00081.safetensors", + "model.layers.43.block_sparse_moe.experts.0.w1.weight": "model-00074-of-00081.safetensors", + "model.layers.43.block_sparse_moe.experts.1.w1.weight": "model-00074-of-00081.safetensors", + "model.layers.43.block_sparse_moe.experts.0.w2.weight": "model-00074-of-00081.safetensors", + "model.layers.43.block_sparse_moe.experts.1.w2.weight": "model-00074-of-00081.safetensors", + "model.layers.44.input_layernorm.weight": "model-00074-of-00081.safetensors", + "model.layers.44.self_attn.q_proj.weight": "model-00074-of-00081.safetensors", + "model.layers.44.self_attn.k_proj.weight": "model-00074-of-00081.safetensors", + "model.layers.44.self_attn.v_proj.weight": "model-00074-of-00081.safetensors", + "model.layers.44.self_attn.o_proj.weight": "model-00075-of-00081.safetensors", + "model.layers.44.post_attention_layernorm.weight": "model-00075-of-00081.safetensors", + "model.layers.44.block_sparse_moe.experts.0.w3.weight": "model-00075-of-00081.safetensors", + "model.layers.44.block_sparse_moe.experts.1.w3.weight": "model-00075-of-00081.safetensors", + "model.layers.44.block_sparse_moe.experts.0.w1.weight": "model-00075-of-00081.safetensors", + "model.layers.44.block_sparse_moe.experts.1.w1.weight": "model-00075-of-00081.safetensors", + "model.layers.44.block_sparse_moe.experts.0.w2.weight": "model-00076-of-00081.safetensors", + "model.layers.44.block_sparse_moe.experts.1.w2.weight": "model-00076-of-00081.safetensors", + "model.layers.45.input_layernorm.weight": "model-00076-of-00081.safetensors", + "model.layers.45.self_attn.q_proj.weight": "model-00076-of-00081.safetensors", + "model.layers.45.self_attn.k_proj.weight": "model-00076-of-00081.safetensors", + "model.layers.45.self_attn.v_proj.weight": "model-00076-of-00081.safetensors", + "model.layers.45.self_attn.o_proj.weight": "model-00076-of-00081.safetensors", + "model.layers.45.post_attention_layernorm.weight": "model-00076-of-00081.safetensors", + "model.layers.45.block_sparse_moe.experts.0.w3.weight": "model-00076-of-00081.safetensors", + "model.layers.45.block_sparse_moe.experts.1.w3.weight": "model-00077-of-00081.safetensors", + "model.layers.45.block_sparse_moe.experts.0.w1.weight": "model-00077-of-00081.safetensors", + "model.layers.45.block_sparse_moe.experts.1.w1.weight": "model-00077-of-00081.safetensors", + "model.layers.45.block_sparse_moe.experts.0.w2.weight": "model-00077-of-00081.safetensors", + "model.layers.45.block_sparse_moe.experts.1.w2.weight": "model-00078-of-00081.safetensors", + "model.layers.46.input_layernorm.weight": "model-00078-of-00081.safetensors", + "model.layers.46.self_attn.q_proj.weight": "model-00078-of-00081.safetensors", + "model.layers.46.self_attn.k_proj.weight": "model-00078-of-00081.safetensors", + "model.layers.46.self_attn.v_proj.weight": "model-00078-of-00081.safetensors", + "model.layers.46.self_attn.o_proj.weight": "model-00078-of-00081.safetensors", + "model.layers.46.post_attention_layernorm.weight": "model-00078-of-00081.safetensors", + "model.layers.46.block_sparse_moe.experts.0.w3.weight": "model-00078-of-00081.safetensors", + "model.layers.46.block_sparse_moe.experts.1.w3.weight": "model-00078-of-00081.safetensors", + "model.layers.46.block_sparse_moe.experts.0.w1.weight": "model-00079-of-00081.safetensors", + "model.layers.46.block_sparse_moe.experts.1.w1.weight": "model-00079-of-00081.safetensors", + "model.layers.46.block_sparse_moe.experts.0.w2.weight": "model-00079-of-00081.safetensors", + "model.layers.46.block_sparse_moe.experts.1.w2.weight": "model-00079-of-00081.safetensors", + "model.layers.47.input_layernorm.weight": "model-00079-of-00081.safetensors", + "model.layers.47.self_attn.q_proj.weight": "model-00079-of-00081.safetensors", + "model.layers.47.self_attn.k_proj.weight": "model-00079-of-00081.safetensors", + "model.layers.47.self_attn.v_proj.weight": "model-00079-of-00081.safetensors", + "model.layers.47.self_attn.o_proj.weight": "model-00080-of-00081.safetensors", + "model.layers.47.post_attention_layernorm.weight": "model-00080-of-00081.safetensors", + "model.layers.47.block_sparse_moe.experts.0.w3.weight": "model-00080-of-00081.safetensors", + "model.layers.47.block_sparse_moe.experts.1.w3.weight": "model-00080-of-00081.safetensors", + "model.layers.47.block_sparse_moe.experts.0.w1.weight": "model-00080-of-00081.safetensors", + "model.layers.47.block_sparse_moe.experts.1.w1.weight": "model-00080-of-00081.safetensors", + "model.layers.47.block_sparse_moe.experts.0.w2.weight": "model-00081-of-00081.safetensors", + "model.layers.47.block_sparse_moe.experts.1.w2.weight": "model-00081-of-00081.safetensors", + "model.norm.weight": "model-00081-of-00081.safetensors", + "lm_head.weight": "model-00081-of-00081.safetensors", + "model.layers.0.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.1.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.2.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.3.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.4.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.5.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.6.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.7.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.8.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.9.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.10.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.11.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.12.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.13.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.14.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.15.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.16.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.17.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.18.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.19.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.20.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.21.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.22.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.23.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.24.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.25.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.26.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.27.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.28.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.29.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.30.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.31.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.32.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.33.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.34.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.35.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.36.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.37.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.38.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.39.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.40.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.41.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.42.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.43.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.44.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.45.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.46.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors", + "model.layers.47.block_sparse_moe.gate.weight": "model-00081-of-00081.safetensors" + } +} \ No newline at end of file diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..53ee8e944e5b091c5735d74342d9a64d72903000 --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b72100dc774faa22934005a522898aafdf2662b3d0e3610e56857a4ec1d99274 +size 20072579 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..8ce2ab04cf0ae124ffd4e7461ae2c1a0bed2d73a --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,31 @@ +{ + "backend": "tokenizers", + "bos_token": "", + "eos_token": "", + "extra_special_tokens": [ + "", + "", + "[BOS]", + "[EOS]", + "<|end_of_msg|>", + "<|open|>", + "<|close|>", + "<|sep|>", + "[start_header_id]", + "[end_header_id]", + "[EOT]", + "<|media_begin|>", + "<|media_content|>", + "<|media_end|>", + "<|media_pad|>", + "", + "[UNK]", + "[PAD]" + ], + "is_local": true, + "local_files_only": false, + "model_max_length": 1000000000000000019884624838656, + "pad_token": "", + "tokenizer_class": "TokenizersBackend", + "unk_token": "" +}