Upload RWForCausalLM

config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "tiiuae/falcon-40B",
+  "alibi": false,
+  "apply_residual_connection_post_layernorm": false,
+  "architectures": [
+    "RWForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "tiiuae/falcon-40B--configuration_RW.RWConfig",
+    "AutoModel": "tiiuae/falcon-40B--modelling_RW.RWModel",
+    "AutoModelForCausalLM": "tiiuae/falcon-40B--modelling_RW.RWForCausalLM",
+    "AutoModelForQuestionAnswering": "tiiuae/falcon-40B--modelling_RW.RWForQuestionAnswering",
+    "AutoModelForSequenceClassification": "tiiuae/falcon-40B--modelling_RW.RWForSequenceClassification",
+    "AutoModelForTokenClassification": "tiiuae/falcon-40B--modelling_RW.RWForTokenClassification"
+  },
+  "bias": false,
+  "bos_token_id": 11,
+  "eos_token_id": 11,
+  "hidden_dropout": 0.0,
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "RefinedWeb",
+  "n_head": 128,
+  "n_head_kv": 8,
+  "n_layer": 60,
+  "parallel_attn": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.31.0",
+  "use_cache": true,
+  "vocab_size": 65024
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.31.0"
+}

model-00001-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66a23c42e94bf13ea4ec7e601e721263c53b75ef560bc137ea2de39cbb7761af
+size 3934456216

model-00002-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed817e57c5ac4599828f9ebaffe07b37d5ac8c1b8a55bf65f743a77619971850
+size 3926067768

model-00003-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e49aa853904b13ee8faddca65faafd522512c72cd03741e769ca1dfd4d159e5b
+size 3540125864

model-00004-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20194e53a11cbf043febb6284b40744af4f959487fa768f34fadf0fbe5c646c8
+size 3540191760

model-00005-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b1cad1d2ee8c90981648ef60d08b09a49cf79f7a2678399d4d619556147fe48
+size 3942845016

model-00006-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:351fdee8b0397d6ee19547529a38e37fd4c2a6e3f6b1dd48b9ce581b9781a2b6
+size 3926067784

model-00007-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d3602f1792cabddfb79cb82c61d9779ba63486e3a3862eba798436f9918fcb6
+size 3540125880

model-00008-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fdb74913d7d2f9bb3ece505c3cee61d1f04e388f94e518c2ec55b0a07ec19766
+size 3540191824

model-00009-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:990de7d2c728cbb9fbea0f1b98e80c7e02c0baa194019e94118959f8117608b6
+size 3942845016

model-00010-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd06368e4b5345361baf2bc7f969dea9eb3cb6145fdb84515f07bfaec1612bd3
+size 3926067784

model-00011-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ec6aad41d76fdf11daf15c79e67f09c806ac4d1a28034bb0929ed79add4ffba
+size 3540125880

model-00012-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:458bd358ec8ecbda160f3e682511760eb221b36ce9c6beeb63c0822c7a27afb7
+size 3540191824

model-00013-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b59d37122bdcfabb10e31e16fbc8b1f61bfad0e5995d66932c002489397ef8ca
+size 3942845016

model-00014-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83e5922e7c06491e6cd2b3c3dc486cac57df6dbe5041c352acfd5508d1f3f393
+size 3926067784

model-00015-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ed4ab7898debfe01d4e1c203bc266001473b05bac081dbd97fa6c544ca09f4e
+size 3540125880

model-00016-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f016099dc86dd23cda6a5bee5dd1f60241385c13b271d4c679da3e8019ca2ee
+size 3540191824

model-00017-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee190064125386fc7561b637fbca4de1dc8c464320b3e479f3cf16b1a599ed42
+size 3942845016

model-00018-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1cc38b7f46428cfacc172b050ef1e4c63bb19f3b383d8b0cb5401a58057ba883
+size 3926067784

model-00019-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:111c6728e64cce066a409fc8969ea8aa58ceeaa25ffecdac1798d1e7440a1cb0
+size 3540125880

model-00020-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40302afd7d55ec58a3621b1958bd1fbcd05aec3307f8fbd4e31d170405730faa
+size 3540191824

model-00021-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5a717122daa7eaaf5b8d97b928a843994c9db0d85df0ce45a84163e801b3c17
+size 3942845016

model-00022-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b097ed70d6b34dafc57ecff58e3ebc871ce87739c1e7a50e9137b0585db7368
+size 3926034800

model.safetensors.index.json

@@ -0,0 +1,490 @@
+{
+  "metadata": {
+    "total_size": 82606587904
+  },
+  "weight_map": {
+    "transformer.h.0.ln_attn.bias": "model-00001-of-00022.safetensors",
+    "transformer.h.0.ln_attn.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.0.ln_mlp.bias": "model-00001-of-00022.safetensors",
+    "transformer.h.0.ln_mlp.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.0.mlp.dense_4h_to_h.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.0.mlp.dense_h_to_4h.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.0.self_attention.dense.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.0.self_attention.query_key_value.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.1.ln_attn.bias": "model-00001-of-00022.safetensors",
+    "transformer.h.1.ln_attn.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.1.ln_mlp.bias": "model-00001-of-00022.safetensors",
+    "transformer.h.1.ln_mlp.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.1.mlp.dense_4h_to_h.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.1.mlp.dense_h_to_4h.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.1.self_attention.dense.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.1.self_attention.query_key_value.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.10.ln_attn.bias": "model-00004-of-00022.safetensors",
+    "transformer.h.10.ln_attn.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.10.ln_mlp.bias": "model-00004-of-00022.safetensors",
+    "transformer.h.10.ln_mlp.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.10.mlp.dense_4h_to_h.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.10.mlp.dense_h_to_4h.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.10.self_attention.dense.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.10.self_attention.query_key_value.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.11.ln_attn.bias": "model-00005-of-00022.safetensors",
+    "transformer.h.11.ln_attn.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.11.ln_mlp.bias": "model-00005-of-00022.safetensors",
+    "transformer.h.11.ln_mlp.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.11.mlp.dense_4h_to_h.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.11.mlp.dense_h_to_4h.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.11.self_attention.dense.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.11.self_attention.query_key_value.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.12.ln_attn.bias": "model-00005-of-00022.safetensors",
+    "transformer.h.12.ln_attn.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.12.ln_mlp.bias": "model-00005-of-00022.safetensors",
+    "transformer.h.12.ln_mlp.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.12.mlp.dense_4h_to_h.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.12.mlp.dense_h_to_4h.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.12.self_attention.dense.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.12.self_attention.query_key_value.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.13.ln_attn.bias": "model-00005-of-00022.safetensors",
+    "transformer.h.13.ln_attn.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.13.ln_mlp.bias": "model-00005-of-00022.safetensors",
+    "transformer.h.13.ln_mlp.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.13.mlp.dense_4h_to_h.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.13.mlp.dense_h_to_4h.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.13.self_attention.dense.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.13.self_attention.query_key_value.weight": "model-00005-of-00022.safetensors",
+    "transformer.h.14.ln_attn.bias": "model-00006-of-00022.safetensors",
+    "transformer.h.14.ln_attn.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.14.ln_mlp.bias": "model-00006-of-00022.safetensors",
+    "transformer.h.14.ln_mlp.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.14.mlp.dense_4h_to_h.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.14.mlp.dense_h_to_4h.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.14.self_attention.dense.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.14.self_attention.query_key_value.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.15.ln_attn.bias": "model-00006-of-00022.safetensors",
+    "transformer.h.15.ln_attn.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.15.ln_mlp.bias": "model-00006-of-00022.safetensors",
+    "transformer.h.15.ln_mlp.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.15.mlp.dense_4h_to_h.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.15.mlp.dense_h_to_4h.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.15.self_attention.dense.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.15.self_attention.query_key_value.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.16.ln_attn.bias": "model-00006-of-00022.safetensors",
+    "transformer.h.16.ln_attn.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.16.ln_mlp.bias": "model-00006-of-00022.safetensors",
+    "transformer.h.16.ln_mlp.weight": "model-00006-of-00022.safetensors",
+    "transformer.h.16.mlp.dense_4h_to_h.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.16.mlp.dense_h_to_4h.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.16.self_attention.dense.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.16.self_attention.query_key_value.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.17.ln_attn.bias": "model-00007-of-00022.safetensors",
+    "transformer.h.17.ln_attn.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.17.ln_mlp.bias": "model-00007-of-00022.safetensors",
+    "transformer.h.17.ln_mlp.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.17.mlp.dense_4h_to_h.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.17.mlp.dense_h_to_4h.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.17.self_attention.dense.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.17.self_attention.query_key_value.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.18.ln_attn.bias": "model-00007-of-00022.safetensors",
+    "transformer.h.18.ln_attn.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.18.ln_mlp.bias": "model-00007-of-00022.safetensors",
+    "transformer.h.18.ln_mlp.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.18.mlp.dense_4h_to_h.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.18.mlp.dense_h_to_4h.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.18.self_attention.dense.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.18.self_attention.query_key_value.weight": "model-00007-of-00022.safetensors",
+    "transformer.h.19.ln_attn.bias": "model-00008-of-00022.safetensors",
+    "transformer.h.19.ln_attn.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.19.ln_mlp.bias": "model-00008-of-00022.safetensors",
+    "transformer.h.19.ln_mlp.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.19.mlp.dense_4h_to_h.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.19.mlp.dense_h_to_4h.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.19.self_attention.dense.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.19.self_attention.query_key_value.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.2.ln_attn.bias": "model-00001-of-00022.safetensors",
+    "transformer.h.2.ln_attn.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.2.ln_mlp.bias": "model-00001-of-00022.safetensors",
+    "transformer.h.2.ln_mlp.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.2.mlp.dense_4h_to_h.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.2.mlp.dense_h_to_4h.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.2.self_attention.dense.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.2.self_attention.query_key_value.weight": "model-00001-of-00022.safetensors",
+    "transformer.h.20.ln_attn.bias": "model-00008-of-00022.safetensors",
+    "transformer.h.20.ln_attn.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.20.ln_mlp.bias": "model-00008-of-00022.safetensors",
+    "transformer.h.20.ln_mlp.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.20.mlp.dense_4h_to_h.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.20.mlp.dense_h_to_4h.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.20.self_attention.dense.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.20.self_attention.query_key_value.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.21.ln_attn.bias": "model-00008-of-00022.safetensors",
+    "transformer.h.21.ln_attn.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.21.ln_mlp.bias": "model-00008-of-00022.safetensors",
+    "transformer.h.21.ln_mlp.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.21.mlp.dense_4h_to_h.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.21.mlp.dense_h_to_4h.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.21.self_attention.dense.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.21.self_attention.query_key_value.weight": "model-00008-of-00022.safetensors",
+    "transformer.h.22.ln_attn.bias": "model-00009-of-00022.safetensors",
+    "transformer.h.22.ln_attn.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.22.ln_mlp.bias": "model-00009-of-00022.safetensors",
+    "transformer.h.22.ln_mlp.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.22.mlp.dense_4h_to_h.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.22.mlp.dense_h_to_4h.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.22.self_attention.dense.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.22.self_attention.query_key_value.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.23.ln_attn.bias": "model-00009-of-00022.safetensors",
+    "transformer.h.23.ln_attn.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.23.ln_mlp.bias": "model-00009-of-00022.safetensors",
+    "transformer.h.23.ln_mlp.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.23.mlp.dense_4h_to_h.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.23.mlp.dense_h_to_4h.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.23.self_attention.dense.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.23.self_attention.query_key_value.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.24.ln_attn.bias": "model-00009-of-00022.safetensors",
+    "transformer.h.24.ln_attn.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.24.ln_mlp.bias": "model-00009-of-00022.safetensors",
+    "transformer.h.24.ln_mlp.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.24.mlp.dense_4h_to_h.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.24.mlp.dense_h_to_4h.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.24.self_attention.dense.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.24.self_attention.query_key_value.weight": "model-00009-of-00022.safetensors",
+    "transformer.h.25.ln_attn.bias": "model-00010-of-00022.safetensors",
+    "transformer.h.25.ln_attn.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.25.ln_mlp.bias": "model-00010-of-00022.safetensors",
+    "transformer.h.25.ln_mlp.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.25.mlp.dense_4h_to_h.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.25.mlp.dense_h_to_4h.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.25.self_attention.dense.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.25.self_attention.query_key_value.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.26.ln_attn.bias": "model-00010-of-00022.safetensors",
+    "transformer.h.26.ln_attn.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.26.ln_mlp.bias": "model-00010-of-00022.safetensors",
+    "transformer.h.26.ln_mlp.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.26.mlp.dense_4h_to_h.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.26.mlp.dense_h_to_4h.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.26.self_attention.dense.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.26.self_attention.query_key_value.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.27.ln_attn.bias": "model-00010-of-00022.safetensors",
+    "transformer.h.27.ln_attn.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.27.ln_mlp.bias": "model-00010-of-00022.safetensors",
+    "transformer.h.27.ln_mlp.weight": "model-00010-of-00022.safetensors",
+    "transformer.h.27.mlp.dense_4h_to_h.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.27.mlp.dense_h_to_4h.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.27.self_attention.dense.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.27.self_attention.query_key_value.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.28.ln_attn.bias": "model-00011-of-00022.safetensors",
+    "transformer.h.28.ln_attn.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.28.ln_mlp.bias": "model-00011-of-00022.safetensors",
+    "transformer.h.28.ln_mlp.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.28.mlp.dense_4h_to_h.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.28.mlp.dense_h_to_4h.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.28.self_attention.dense.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.28.self_attention.query_key_value.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.29.ln_attn.bias": "model-00011-of-00022.safetensors",
+    "transformer.h.29.ln_attn.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.29.ln_mlp.bias": "model-00011-of-00022.safetensors",
+    "transformer.h.29.ln_mlp.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.29.mlp.dense_4h_to_h.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.29.mlp.dense_h_to_4h.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.29.self_attention.dense.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.29.self_attention.query_key_value.weight": "model-00011-of-00022.safetensors",
+    "transformer.h.3.ln_attn.bias": "model-00002-of-00022.safetensors",
+    "transformer.h.3.ln_attn.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.3.ln_mlp.bias": "model-00002-of-00022.safetensors",
+    "transformer.h.3.ln_mlp.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.3.mlp.dense_4h_to_h.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.3.mlp.dense_h_to_4h.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.3.self_attention.dense.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.3.self_attention.query_key_value.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.30.ln_attn.bias": "model-00012-of-00022.safetensors",
+    "transformer.h.30.ln_attn.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.30.ln_mlp.bias": "model-00012-of-00022.safetensors",
+    "transformer.h.30.ln_mlp.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.30.mlp.dense_4h_to_h.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.30.mlp.dense_h_to_4h.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.30.self_attention.dense.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.30.self_attention.query_key_value.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.31.ln_attn.bias": "model-00012-of-00022.safetensors",
+    "transformer.h.31.ln_attn.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.31.ln_mlp.bias": "model-00012-of-00022.safetensors",
+    "transformer.h.31.ln_mlp.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.31.mlp.dense_4h_to_h.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.31.mlp.dense_h_to_4h.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.31.self_attention.dense.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.31.self_attention.query_key_value.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.32.ln_attn.bias": "model-00012-of-00022.safetensors",
+    "transformer.h.32.ln_attn.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.32.ln_mlp.bias": "model-00012-of-00022.safetensors",
+    "transformer.h.32.ln_mlp.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.32.mlp.dense_4h_to_h.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.32.mlp.dense_h_to_4h.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.32.self_attention.dense.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.32.self_attention.query_key_value.weight": "model-00012-of-00022.safetensors",
+    "transformer.h.33.ln_attn.bias": "model-00013-of-00022.safetensors",
+    "transformer.h.33.ln_attn.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.33.ln_mlp.bias": "model-00013-of-00022.safetensors",
+    "transformer.h.33.ln_mlp.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.33.mlp.dense_4h_to_h.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.33.mlp.dense_h_to_4h.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.33.self_attention.dense.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.33.self_attention.query_key_value.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.34.ln_attn.bias": "model-00013-of-00022.safetensors",
+    "transformer.h.34.ln_attn.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.34.ln_mlp.bias": "model-00013-of-00022.safetensors",
+    "transformer.h.34.ln_mlp.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.34.mlp.dense_4h_to_h.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.34.mlp.dense_h_to_4h.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.34.self_attention.dense.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.34.self_attention.query_key_value.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.35.ln_attn.bias": "model-00013-of-00022.safetensors",
+    "transformer.h.35.ln_attn.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.35.ln_mlp.bias": "model-00013-of-00022.safetensors",
+    "transformer.h.35.ln_mlp.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.35.mlp.dense_4h_to_h.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.35.mlp.dense_h_to_4h.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.35.self_attention.dense.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.35.self_attention.query_key_value.weight": "model-00013-of-00022.safetensors",
+    "transformer.h.36.ln_attn.bias": "model-00014-of-00022.safetensors",
+    "transformer.h.36.ln_attn.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.36.ln_mlp.bias": "model-00014-of-00022.safetensors",
+    "transformer.h.36.ln_mlp.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.36.mlp.dense_4h_to_h.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.36.mlp.dense_h_to_4h.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.36.self_attention.dense.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.36.self_attention.query_key_value.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.37.ln_attn.bias": "model-00014-of-00022.safetensors",
+    "transformer.h.37.ln_attn.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.37.ln_mlp.bias": "model-00014-of-00022.safetensors",
+    "transformer.h.37.ln_mlp.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.37.mlp.dense_4h_to_h.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.37.mlp.dense_h_to_4h.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.37.self_attention.dense.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.37.self_attention.query_key_value.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.38.ln_attn.bias": "model-00014-of-00022.safetensors",
+    "transformer.h.38.ln_attn.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.38.ln_mlp.bias": "model-00014-of-00022.safetensors",
+    "transformer.h.38.ln_mlp.weight": "model-00014-of-00022.safetensors",
+    "transformer.h.38.mlp.dense_4h_to_h.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.38.mlp.dense_h_to_4h.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.38.self_attention.dense.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.38.self_attention.query_key_value.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.39.ln_attn.bias": "model-00015-of-00022.safetensors",
+    "transformer.h.39.ln_attn.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.39.ln_mlp.bias": "model-00015-of-00022.safetensors",
+    "transformer.h.39.ln_mlp.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.39.mlp.dense_4h_to_h.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.39.mlp.dense_h_to_4h.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.39.self_attention.dense.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.39.self_attention.query_key_value.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.4.ln_attn.bias": "model-00002-of-00022.safetensors",
+    "transformer.h.4.ln_attn.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.4.ln_mlp.bias": "model-00002-of-00022.safetensors",
+    "transformer.h.4.ln_mlp.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.4.mlp.dense_4h_to_h.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.4.mlp.dense_h_to_4h.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.4.self_attention.dense.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.4.self_attention.query_key_value.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.40.ln_attn.bias": "model-00015-of-00022.safetensors",
+    "transformer.h.40.ln_attn.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.40.ln_mlp.bias": "model-00015-of-00022.safetensors",
+    "transformer.h.40.ln_mlp.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.40.mlp.dense_4h_to_h.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.40.mlp.dense_h_to_4h.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.40.self_attention.dense.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.40.self_attention.query_key_value.weight": "model-00015-of-00022.safetensors",
+    "transformer.h.41.ln_attn.bias": "model-00016-of-00022.safetensors",
+    "transformer.h.41.ln_attn.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.41.ln_mlp.bias": "model-00016-of-00022.safetensors",
+    "transformer.h.41.ln_mlp.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.41.mlp.dense_4h_to_h.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.41.mlp.dense_h_to_4h.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.41.self_attention.dense.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.41.self_attention.query_key_value.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.42.ln_attn.bias": "model-00016-of-00022.safetensors",
+    "transformer.h.42.ln_attn.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.42.ln_mlp.bias": "model-00016-of-00022.safetensors",
+    "transformer.h.42.ln_mlp.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.42.mlp.dense_4h_to_h.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.42.mlp.dense_h_to_4h.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.42.self_attention.dense.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.42.self_attention.query_key_value.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.43.ln_attn.bias": "model-00016-of-00022.safetensors",
+    "transformer.h.43.ln_attn.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.43.ln_mlp.bias": "model-00016-of-00022.safetensors",
+    "transformer.h.43.ln_mlp.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.43.mlp.dense_4h_to_h.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.43.mlp.dense_h_to_4h.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.43.self_attention.dense.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.43.self_attention.query_key_value.weight": "model-00016-of-00022.safetensors",
+    "transformer.h.44.ln_attn.bias": "model-00017-of-00022.safetensors",
+    "transformer.h.44.ln_attn.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.44.ln_mlp.bias": "model-00017-of-00022.safetensors",
+    "transformer.h.44.ln_mlp.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.44.mlp.dense_4h_to_h.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.44.mlp.dense_h_to_4h.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.44.self_attention.dense.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.44.self_attention.query_key_value.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.45.ln_attn.bias": "model-00017-of-00022.safetensors",
+    "transformer.h.45.ln_attn.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.45.ln_mlp.bias": "model-00017-of-00022.safetensors",
+    "transformer.h.45.ln_mlp.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.45.mlp.dense_4h_to_h.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.45.mlp.dense_h_to_4h.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.45.self_attention.dense.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.45.self_attention.query_key_value.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.46.ln_attn.bias": "model-00017-of-00022.safetensors",
+    "transformer.h.46.ln_attn.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.46.ln_mlp.bias": "model-00017-of-00022.safetensors",
+    "transformer.h.46.ln_mlp.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.46.mlp.dense_4h_to_h.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.46.mlp.dense_h_to_4h.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.46.self_attention.dense.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.46.self_attention.query_key_value.weight": "model-00017-of-00022.safetensors",
+    "transformer.h.47.ln_attn.bias": "model-00018-of-00022.safetensors",
+    "transformer.h.47.ln_attn.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.47.ln_mlp.bias": "model-00018-of-00022.safetensors",
+    "transformer.h.47.ln_mlp.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.47.mlp.dense_4h_to_h.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.47.mlp.dense_h_to_4h.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.47.self_attention.dense.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.47.self_attention.query_key_value.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.48.ln_attn.bias": "model-00018-of-00022.safetensors",
+    "transformer.h.48.ln_attn.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.48.ln_mlp.bias": "model-00018-of-00022.safetensors",
+    "transformer.h.48.ln_mlp.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.48.mlp.dense_4h_to_h.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.48.mlp.dense_h_to_4h.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.48.self_attention.dense.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.48.self_attention.query_key_value.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.49.ln_attn.bias": "model-00018-of-00022.safetensors",
+    "transformer.h.49.ln_attn.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.49.ln_mlp.bias": "model-00018-of-00022.safetensors",
+    "transformer.h.49.ln_mlp.weight": "model-00018-of-00022.safetensors",
+    "transformer.h.49.mlp.dense_4h_to_h.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.49.mlp.dense_h_to_4h.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.49.self_attention.dense.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.49.self_attention.query_key_value.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.5.ln_attn.bias": "model-00002-of-00022.safetensors",
+    "transformer.h.5.ln_attn.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.5.ln_mlp.bias": "model-00002-of-00022.safetensors",
+    "transformer.h.5.ln_mlp.weight": "model-00002-of-00022.safetensors",
+    "transformer.h.5.mlp.dense_4h_to_h.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.5.mlp.dense_h_to_4h.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.5.self_attention.dense.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.5.self_attention.query_key_value.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.50.ln_attn.bias": "model-00019-of-00022.safetensors",
+    "transformer.h.50.ln_attn.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.50.ln_mlp.bias": "model-00019-of-00022.safetensors",
+    "transformer.h.50.ln_mlp.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.50.mlp.dense_4h_to_h.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.50.mlp.dense_h_to_4h.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.50.self_attention.dense.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.50.self_attention.query_key_value.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.51.ln_attn.bias": "model-00019-of-00022.safetensors",
+    "transformer.h.51.ln_attn.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.51.ln_mlp.bias": "model-00019-of-00022.safetensors",
+    "transformer.h.51.ln_mlp.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.51.mlp.dense_4h_to_h.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.51.mlp.dense_h_to_4h.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.51.self_attention.dense.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.51.self_attention.query_key_value.weight": "model-00019-of-00022.safetensors",
+    "transformer.h.52.ln_attn.bias": "model-00020-of-00022.safetensors",
+    "transformer.h.52.ln_attn.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.52.ln_mlp.bias": "model-00020-of-00022.safetensors",
+    "transformer.h.52.ln_mlp.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.52.mlp.dense_4h_to_h.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.52.mlp.dense_h_to_4h.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.52.self_attention.dense.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.52.self_attention.query_key_value.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.53.ln_attn.bias": "model-00020-of-00022.safetensors",
+    "transformer.h.53.ln_attn.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.53.ln_mlp.bias": "model-00020-of-00022.safetensors",
+    "transformer.h.53.ln_mlp.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.53.mlp.dense_4h_to_h.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.53.mlp.dense_h_to_4h.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.53.self_attention.dense.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.53.self_attention.query_key_value.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.54.ln_attn.bias": "model-00020-of-00022.safetensors",
+    "transformer.h.54.ln_attn.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.54.ln_mlp.bias": "model-00020-of-00022.safetensors",
+    "transformer.h.54.ln_mlp.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.54.mlp.dense_4h_to_h.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.54.mlp.dense_h_to_4h.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.54.self_attention.dense.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.54.self_attention.query_key_value.weight": "model-00020-of-00022.safetensors",
+    "transformer.h.55.ln_attn.bias": "model-00021-of-00022.safetensors",
+    "transformer.h.55.ln_attn.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.55.ln_mlp.bias": "model-00021-of-00022.safetensors",
+    "transformer.h.55.ln_mlp.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.55.mlp.dense_4h_to_h.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.55.mlp.dense_h_to_4h.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.55.self_attention.dense.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.55.self_attention.query_key_value.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.56.ln_attn.bias": "model-00021-of-00022.safetensors",
+    "transformer.h.56.ln_attn.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.56.ln_mlp.bias": "model-00021-of-00022.safetensors",
+    "transformer.h.56.ln_mlp.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.56.mlp.dense_4h_to_h.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.56.mlp.dense_h_to_4h.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.56.self_attention.dense.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.56.self_attention.query_key_value.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.57.ln_attn.bias": "model-00021-of-00022.safetensors",
+    "transformer.h.57.ln_attn.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.57.ln_mlp.bias": "model-00021-of-00022.safetensors",
+    "transformer.h.57.ln_mlp.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.57.mlp.dense_4h_to_h.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.57.mlp.dense_h_to_4h.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.57.self_attention.dense.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.57.self_attention.query_key_value.weight": "model-00021-of-00022.safetensors",
+    "transformer.h.58.ln_attn.bias": "model-00022-of-00022.safetensors",
+    "transformer.h.58.ln_attn.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.58.ln_mlp.bias": "model-00022-of-00022.safetensors",
+    "transformer.h.58.ln_mlp.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.58.mlp.dense_4h_to_h.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.58.mlp.dense_h_to_4h.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.58.self_attention.dense.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.58.self_attention.query_key_value.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.59.ln_attn.bias": "model-00022-of-00022.safetensors",
+    "transformer.h.59.ln_attn.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.59.ln_mlp.bias": "model-00022-of-00022.safetensors",
+    "transformer.h.59.ln_mlp.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.59.mlp.dense_4h_to_h.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.59.mlp.dense_h_to_4h.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.59.self_attention.dense.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.59.self_attention.query_key_value.weight": "model-00022-of-00022.safetensors",
+    "transformer.h.6.ln_attn.bias": "model-00003-of-00022.safetensors",
+    "transformer.h.6.ln_attn.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.6.ln_mlp.bias": "model-00003-of-00022.safetensors",
+    "transformer.h.6.ln_mlp.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.6.mlp.dense_4h_to_h.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.6.mlp.dense_h_to_4h.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.6.self_attention.dense.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.6.self_attention.query_key_value.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.7.ln_attn.bias": "model-00003-of-00022.safetensors",
+    "transformer.h.7.ln_attn.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.7.ln_mlp.bias": "model-00003-of-00022.safetensors",
+    "transformer.h.7.ln_mlp.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.7.mlp.dense_4h_to_h.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.7.mlp.dense_h_to_4h.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.7.self_attention.dense.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.7.self_attention.query_key_value.weight": "model-00003-of-00022.safetensors",
+    "transformer.h.8.ln_attn.bias": "model-00004-of-00022.safetensors",
+    "transformer.h.8.ln_attn.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.8.ln_mlp.bias": "model-00004-of-00022.safetensors",
+    "transformer.h.8.ln_mlp.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.8.mlp.dense_4h_to_h.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.8.mlp.dense_h_to_4h.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.8.self_attention.dense.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.8.self_attention.query_key_value.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.9.ln_attn.bias": "model-00004-of-00022.safetensors",
+    "transformer.h.9.ln_attn.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.9.ln_mlp.bias": "model-00004-of-00022.safetensors",
+    "transformer.h.9.ln_mlp.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.9.mlp.dense_4h_to_h.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.9.mlp.dense_h_to_4h.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.9.self_attention.dense.weight": "model-00004-of-00022.safetensors",
+    "transformer.h.9.self_attention.query_key_value.weight": "model-00004-of-00022.safetensors",
+    "transformer.ln_f.bias": "model-00022-of-00022.safetensors",
+    "transformer.ln_f.weight": "model-00022-of-00022.safetensors",
+    "transformer.word_embeddings.weight": "model-00001-of-00022.safetensors"
+  }
+}