Upload folder using huggingface_hub
Browse files- model-00001-of-00004.safetensors +3 -0
- model-00002-of-00004.safetensors +3 -0
- model-00003-of-00004.safetensors +3 -0
- model-00004-of-00004.safetensors +3 -0
- optimizer.pt +3 -0
- scheduler.pt +1 -1
- trainer_state.json +353 -3
- training_args.bin +1 -1
model-00001-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:089b26b8d969b2ae77a78b434481eba1aa73b3b569a816259e837318069c7587
|
| 3 |
+
size 4976698672
|
model-00002-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0744e2fd8b4a5877f08b57fea9cfce9d0518b35c48a4b31309d451a349d8171d
|
| 3 |
+
size 4999802720
|
model-00003-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0385113952a34362a3495d2cd125276e2c41470a20bfe3146caab543ce32e1b3
|
| 3 |
+
size 4915916176
|
model-00004-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9e7d609a3b86e54902f265ebe76e6b181636d223bb6f77a425e4178f3e04964
|
| 3 |
+
size 1168138808
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d841224710771e9f8dfb28dff603765300ae4045ca3072f9241252b71014d0c3
|
| 3 |
+
size 32121299754
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e73dde419b037c5cd50ee6571f9536cd16a1e6dbafab43748949db14983e8eca
|
| 3 |
size 1064
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -7007,6 +7007,356 @@
|
|
| 7007 |
"learning_rate": 1.8e-05,
|
| 7008 |
"loss": 5.1117,
|
| 7009 |
"step": 10000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7010 |
}
|
| 7011 |
],
|
| 7012 |
"logging_steps": 100,
|
|
@@ -7014,7 +7364,7 @@
|
|
| 7014 |
"num_input_tokens_seen": 0,
|
| 7015 |
"num_train_epochs": 1,
|
| 7016 |
"save_steps": 5000,
|
| 7017 |
-
"total_flos":
|
| 7018 |
"train_batch_size": 4,
|
| 7019 |
"trial_name": null,
|
| 7020 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.004685567203528545,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 15000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 7007 |
"learning_rate": 1.8e-05,
|
| 7008 |
"loss": 5.1117,
|
| 7009 |
"step": 10000
|
| 7010 |
+
},
|
| 7011 |
+
{
|
| 7012 |
+
"epoch": 0.00315494858370922,
|
| 7013 |
+
"grad_norm": 12.6875,
|
| 7014 |
+
"learning_rate": 1.798e-05,
|
| 7015 |
+
"loss": 5.5697,
|
| 7016 |
+
"step": 10100
|
| 7017 |
+
},
|
| 7018 |
+
{
|
| 7019 |
+
"epoch": 0.00318618569839941,
|
| 7020 |
+
"grad_norm": 17.75,
|
| 7021 |
+
"learning_rate": 1.796e-05,
|
| 7022 |
+
"loss": 5.6353,
|
| 7023 |
+
"step": 10200
|
| 7024 |
+
},
|
| 7025 |
+
{
|
| 7026 |
+
"epoch": 0.0032174228130896006,
|
| 7027 |
+
"grad_norm": 14.1875,
|
| 7028 |
+
"learning_rate": 1.794e-05,
|
| 7029 |
+
"loss": 5.5588,
|
| 7030 |
+
"step": 10300
|
| 7031 |
+
},
|
| 7032 |
+
{
|
| 7033 |
+
"epoch": 0.003248659927779791,
|
| 7034 |
+
"grad_norm": 21.0,
|
| 7035 |
+
"learning_rate": 1.792e-05,
|
| 7036 |
+
"loss": 5.5504,
|
| 7037 |
+
"step": 10400
|
| 7038 |
+
},
|
| 7039 |
+
{
|
| 7040 |
+
"epoch": 0.003279897042469981,
|
| 7041 |
+
"grad_norm": 20.5,
|
| 7042 |
+
"learning_rate": 1.79e-05,
|
| 7043 |
+
"loss": 5.5269,
|
| 7044 |
+
"step": 10500
|
| 7045 |
+
},
|
| 7046 |
+
{
|
| 7047 |
+
"epoch": 0.0033111341571601715,
|
| 7048 |
+
"grad_norm": 20.625,
|
| 7049 |
+
"learning_rate": 1.788e-05,
|
| 7050 |
+
"loss": 5.4473,
|
| 7051 |
+
"step": 10600
|
| 7052 |
+
},
|
| 7053 |
+
{
|
| 7054 |
+
"epoch": 0.003342371271850362,
|
| 7055 |
+
"grad_norm": 13.4375,
|
| 7056 |
+
"learning_rate": 1.7860000000000002e-05,
|
| 7057 |
+
"loss": 5.5347,
|
| 7058 |
+
"step": 10700
|
| 7059 |
+
},
|
| 7060 |
+
{
|
| 7061 |
+
"epoch": 0.003373608386540552,
|
| 7062 |
+
"grad_norm": 17.625,
|
| 7063 |
+
"learning_rate": 1.7840000000000002e-05,
|
| 7064 |
+
"loss": 5.4918,
|
| 7065 |
+
"step": 10800
|
| 7066 |
+
},
|
| 7067 |
+
{
|
| 7068 |
+
"epoch": 0.0034048455012307423,
|
| 7069 |
+
"grad_norm": 17.75,
|
| 7070 |
+
"learning_rate": 1.7820000000000002e-05,
|
| 7071 |
+
"loss": 5.5585,
|
| 7072 |
+
"step": 10900
|
| 7073 |
+
},
|
| 7074 |
+
{
|
| 7075 |
+
"epoch": 0.0034360826159209328,
|
| 7076 |
+
"grad_norm": 20.125,
|
| 7077 |
+
"learning_rate": 1.7800000000000002e-05,
|
| 7078 |
+
"loss": 5.578,
|
| 7079 |
+
"step": 11000
|
| 7080 |
+
},
|
| 7081 |
+
{
|
| 7082 |
+
"epoch": 0.0034673197306111228,
|
| 7083 |
+
"grad_norm": 17.25,
|
| 7084 |
+
"learning_rate": 1.7780000000000003e-05,
|
| 7085 |
+
"loss": 5.5814,
|
| 7086 |
+
"step": 11100
|
| 7087 |
+
},
|
| 7088 |
+
{
|
| 7089 |
+
"epoch": 0.003498556845301313,
|
| 7090 |
+
"grad_norm": 18.25,
|
| 7091 |
+
"learning_rate": 1.7760000000000003e-05,
|
| 7092 |
+
"loss": 5.5033,
|
| 7093 |
+
"step": 11200
|
| 7094 |
+
},
|
| 7095 |
+
{
|
| 7096 |
+
"epoch": 0.0035297939599915036,
|
| 7097 |
+
"grad_norm": 12.8125,
|
| 7098 |
+
"learning_rate": 1.7740000000000003e-05,
|
| 7099 |
+
"loss": 5.5739,
|
| 7100 |
+
"step": 11300
|
| 7101 |
+
},
|
| 7102 |
+
{
|
| 7103 |
+
"epoch": 0.0035610310746816936,
|
| 7104 |
+
"grad_norm": 13.75,
|
| 7105 |
+
"learning_rate": 1.7720000000000003e-05,
|
| 7106 |
+
"loss": 5.5183,
|
| 7107 |
+
"step": 11400
|
| 7108 |
+
},
|
| 7109 |
+
{
|
| 7110 |
+
"epoch": 0.003592268189371884,
|
| 7111 |
+
"grad_norm": 17.375,
|
| 7112 |
+
"learning_rate": 1.77e-05,
|
| 7113 |
+
"loss": 5.58,
|
| 7114 |
+
"step": 11500
|
| 7115 |
+
},
|
| 7116 |
+
{
|
| 7117 |
+
"epoch": 0.0036235053040620745,
|
| 7118 |
+
"grad_norm": 15.875,
|
| 7119 |
+
"learning_rate": 1.768e-05,
|
| 7120 |
+
"loss": 5.6626,
|
| 7121 |
+
"step": 11600
|
| 7122 |
+
},
|
| 7123 |
+
{
|
| 7124 |
+
"epoch": 0.0036547424187522645,
|
| 7125 |
+
"grad_norm": 15.5,
|
| 7126 |
+
"learning_rate": 1.766e-05,
|
| 7127 |
+
"loss": 5.4935,
|
| 7128 |
+
"step": 11700
|
| 7129 |
+
},
|
| 7130 |
+
{
|
| 7131 |
+
"epoch": 0.003685979533442455,
|
| 7132 |
+
"grad_norm": 17.625,
|
| 7133 |
+
"learning_rate": 1.764e-05,
|
| 7134 |
+
"loss": 5.516,
|
| 7135 |
+
"step": 11800
|
| 7136 |
+
},
|
| 7137 |
+
{
|
| 7138 |
+
"epoch": 0.0037172166481326454,
|
| 7139 |
+
"grad_norm": 14.75,
|
| 7140 |
+
"learning_rate": 1.762e-05,
|
| 7141 |
+
"loss": 5.564,
|
| 7142 |
+
"step": 11900
|
| 7143 |
+
},
|
| 7144 |
+
{
|
| 7145 |
+
"epoch": 0.0037484537628228354,
|
| 7146 |
+
"grad_norm": 16.125,
|
| 7147 |
+
"learning_rate": 1.76e-05,
|
| 7148 |
+
"loss": 5.5759,
|
| 7149 |
+
"step": 12000
|
| 7150 |
+
},
|
| 7151 |
+
{
|
| 7152 |
+
"epoch": 0.003779690877513026,
|
| 7153 |
+
"grad_norm": 18.25,
|
| 7154 |
+
"learning_rate": 1.758e-05,
|
| 7155 |
+
"loss": 5.5458,
|
| 7156 |
+
"step": 12100
|
| 7157 |
+
},
|
| 7158 |
+
{
|
| 7159 |
+
"epoch": 0.0038109279922032163,
|
| 7160 |
+
"grad_norm": 20.75,
|
| 7161 |
+
"learning_rate": 1.756e-05,
|
| 7162 |
+
"loss": 5.6255,
|
| 7163 |
+
"step": 12200
|
| 7164 |
+
},
|
| 7165 |
+
{
|
| 7166 |
+
"epoch": 0.0038421651068934063,
|
| 7167 |
+
"grad_norm": 15.5,
|
| 7168 |
+
"learning_rate": 1.754e-05,
|
| 7169 |
+
"loss": 5.5217,
|
| 7170 |
+
"step": 12300
|
| 7171 |
+
},
|
| 7172 |
+
{
|
| 7173 |
+
"epoch": 0.0038734022215835967,
|
| 7174 |
+
"grad_norm": 15.3125,
|
| 7175 |
+
"learning_rate": 1.752e-05,
|
| 7176 |
+
"loss": 5.5526,
|
| 7177 |
+
"step": 12400
|
| 7178 |
+
},
|
| 7179 |
+
{
|
| 7180 |
+
"epoch": 0.003904639336273787,
|
| 7181 |
+
"grad_norm": 18.625,
|
| 7182 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 7183 |
+
"loss": 5.5548,
|
| 7184 |
+
"step": 12500
|
| 7185 |
+
},
|
| 7186 |
+
{
|
| 7187 |
+
"epoch": 0.003935876450963978,
|
| 7188 |
+
"grad_norm": 15.4375,
|
| 7189 |
+
"learning_rate": 1.7480000000000002e-05,
|
| 7190 |
+
"loss": 5.5379,
|
| 7191 |
+
"step": 12600
|
| 7192 |
+
},
|
| 7193 |
+
{
|
| 7194 |
+
"epoch": 0.003967113565654168,
|
| 7195 |
+
"grad_norm": 19.0,
|
| 7196 |
+
"learning_rate": 1.7460000000000002e-05,
|
| 7197 |
+
"loss": 5.5492,
|
| 7198 |
+
"step": 12700
|
| 7199 |
+
},
|
| 7200 |
+
{
|
| 7201 |
+
"epoch": 0.003998350680344358,
|
| 7202 |
+
"grad_norm": 16.875,
|
| 7203 |
+
"learning_rate": 1.7440000000000002e-05,
|
| 7204 |
+
"loss": 5.5787,
|
| 7205 |
+
"step": 12800
|
| 7206 |
+
},
|
| 7207 |
+
{
|
| 7208 |
+
"epoch": 0.004029587795034548,
|
| 7209 |
+
"grad_norm": 10.3125,
|
| 7210 |
+
"learning_rate": 1.7420000000000003e-05,
|
| 7211 |
+
"loss": 5.4924,
|
| 7212 |
+
"step": 12900
|
| 7213 |
+
},
|
| 7214 |
+
{
|
| 7215 |
+
"epoch": 0.0040608249097247385,
|
| 7216 |
+
"grad_norm": 14.375,
|
| 7217 |
+
"learning_rate": 1.7400000000000003e-05,
|
| 7218 |
+
"loss": 5.4351,
|
| 7219 |
+
"step": 13000
|
| 7220 |
+
},
|
| 7221 |
+
{
|
| 7222 |
+
"epoch": 0.004092062024414929,
|
| 7223 |
+
"grad_norm": 17.25,
|
| 7224 |
+
"learning_rate": 1.7380000000000003e-05,
|
| 7225 |
+
"loss": 5.5044,
|
| 7226 |
+
"step": 13100
|
| 7227 |
+
},
|
| 7228 |
+
{
|
| 7229 |
+
"epoch": 0.004123299139105119,
|
| 7230 |
+
"grad_norm": 13.9375,
|
| 7231 |
+
"learning_rate": 1.736e-05,
|
| 7232 |
+
"loss": 5.5039,
|
| 7233 |
+
"step": 13200
|
| 7234 |
+
},
|
| 7235 |
+
{
|
| 7236 |
+
"epoch": 0.00415453625379531,
|
| 7237 |
+
"grad_norm": 15.4375,
|
| 7238 |
+
"learning_rate": 1.734e-05,
|
| 7239 |
+
"loss": 5.5401,
|
| 7240 |
+
"step": 13300
|
| 7241 |
+
},
|
| 7242 |
+
{
|
| 7243 |
+
"epoch": 0.004185773368485499,
|
| 7244 |
+
"grad_norm": 15.375,
|
| 7245 |
+
"learning_rate": 1.732e-05,
|
| 7246 |
+
"loss": 5.6314,
|
| 7247 |
+
"step": 13400
|
| 7248 |
+
},
|
| 7249 |
+
{
|
| 7250 |
+
"epoch": 0.00421701048317569,
|
| 7251 |
+
"grad_norm": 13.875,
|
| 7252 |
+
"learning_rate": 1.73e-05,
|
| 7253 |
+
"loss": 5.6265,
|
| 7254 |
+
"step": 13500
|
| 7255 |
+
},
|
| 7256 |
+
{
|
| 7257 |
+
"epoch": 0.00424824759786588,
|
| 7258 |
+
"grad_norm": 14.375,
|
| 7259 |
+
"learning_rate": 1.728e-05,
|
| 7260 |
+
"loss": 5.6575,
|
| 7261 |
+
"step": 13600
|
| 7262 |
+
},
|
| 7263 |
+
{
|
| 7264 |
+
"epoch": 0.004279484712556071,
|
| 7265 |
+
"grad_norm": 13.125,
|
| 7266 |
+
"learning_rate": 1.726e-05,
|
| 7267 |
+
"loss": 5.5269,
|
| 7268 |
+
"step": 13700
|
| 7269 |
+
},
|
| 7270 |
+
{
|
| 7271 |
+
"epoch": 0.004310721827246261,
|
| 7272 |
+
"grad_norm": 16.375,
|
| 7273 |
+
"learning_rate": 1.724e-05,
|
| 7274 |
+
"loss": 5.5832,
|
| 7275 |
+
"step": 13800
|
| 7276 |
+
},
|
| 7277 |
+
{
|
| 7278 |
+
"epoch": 0.0043419589419364515,
|
| 7279 |
+
"grad_norm": 15.25,
|
| 7280 |
+
"learning_rate": 1.722e-05,
|
| 7281 |
+
"loss": 5.5344,
|
| 7282 |
+
"step": 13900
|
| 7283 |
+
},
|
| 7284 |
+
{
|
| 7285 |
+
"epoch": 0.004373196056626641,
|
| 7286 |
+
"grad_norm": 13.875,
|
| 7287 |
+
"learning_rate": 1.72e-05,
|
| 7288 |
+
"loss": 5.6058,
|
| 7289 |
+
"step": 14000
|
| 7290 |
+
},
|
| 7291 |
+
{
|
| 7292 |
+
"epoch": 0.0044044331713168315,
|
| 7293 |
+
"grad_norm": 13.25,
|
| 7294 |
+
"learning_rate": 1.718e-05,
|
| 7295 |
+
"loss": 5.5807,
|
| 7296 |
+
"step": 14100
|
| 7297 |
+
},
|
| 7298 |
+
{
|
| 7299 |
+
"epoch": 0.004435670286007022,
|
| 7300 |
+
"grad_norm": 15.4375,
|
| 7301 |
+
"learning_rate": 1.7160000000000002e-05,
|
| 7302 |
+
"loss": 5.569,
|
| 7303 |
+
"step": 14200
|
| 7304 |
+
},
|
| 7305 |
+
{
|
| 7306 |
+
"epoch": 0.004466907400697212,
|
| 7307 |
+
"grad_norm": 15.625,
|
| 7308 |
+
"learning_rate": 1.7140000000000002e-05,
|
| 7309 |
+
"loss": 5.5629,
|
| 7310 |
+
"step": 14300
|
| 7311 |
+
},
|
| 7312 |
+
{
|
| 7313 |
+
"epoch": 0.004498144515387403,
|
| 7314 |
+
"grad_norm": 14.3125,
|
| 7315 |
+
"learning_rate": 1.7120000000000002e-05,
|
| 7316 |
+
"loss": 5.4952,
|
| 7317 |
+
"step": 14400
|
| 7318 |
+
},
|
| 7319 |
+
{
|
| 7320 |
+
"epoch": 0.004529381630077593,
|
| 7321 |
+
"grad_norm": 12.25,
|
| 7322 |
+
"learning_rate": 1.7100000000000002e-05,
|
| 7323 |
+
"loss": 5.567,
|
| 7324 |
+
"step": 14500
|
| 7325 |
+
},
|
| 7326 |
+
{
|
| 7327 |
+
"epoch": 0.004560618744767784,
|
| 7328 |
+
"grad_norm": 15.5625,
|
| 7329 |
+
"learning_rate": 1.7080000000000002e-05,
|
| 7330 |
+
"loss": 5.6412,
|
| 7331 |
+
"step": 14600
|
| 7332 |
+
},
|
| 7333 |
+
{
|
| 7334 |
+
"epoch": 0.004591855859457973,
|
| 7335 |
+
"grad_norm": 12.8125,
|
| 7336 |
+
"learning_rate": 1.7060000000000003e-05,
|
| 7337 |
+
"loss": 5.548,
|
| 7338 |
+
"step": 14700
|
| 7339 |
+
},
|
| 7340 |
+
{
|
| 7341 |
+
"epoch": 0.004623092974148164,
|
| 7342 |
+
"grad_norm": 18.875,
|
| 7343 |
+
"learning_rate": 1.704e-05,
|
| 7344 |
+
"loss": 5.5444,
|
| 7345 |
+
"step": 14800
|
| 7346 |
+
},
|
| 7347 |
+
{
|
| 7348 |
+
"epoch": 0.004654330088838354,
|
| 7349 |
+
"grad_norm": 14.1875,
|
| 7350 |
+
"learning_rate": 1.702e-05,
|
| 7351 |
+
"loss": 5.5596,
|
| 7352 |
+
"step": 14900
|
| 7353 |
+
},
|
| 7354 |
+
{
|
| 7355 |
+
"epoch": 0.004685567203528545,
|
| 7356 |
+
"grad_norm": 11.125,
|
| 7357 |
+
"learning_rate": 1.7e-05,
|
| 7358 |
+
"loss": 5.6135,
|
| 7359 |
+
"step": 15000
|
| 7360 |
}
|
| 7361 |
],
|
| 7362 |
"logging_steps": 100,
|
|
|
|
| 7364 |
"num_input_tokens_seen": 0,
|
| 7365 |
"num_train_epochs": 1,
|
| 7366 |
"save_steps": 5000,
|
| 7367 |
+
"total_flos": 2.028851316129792e+18,
|
| 7368 |
"train_batch_size": 4,
|
| 7369 |
"trial_name": null,
|
| 7370 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1568c6c7d3d9c016e240768e821a0fdc1b961d37139cdac4ce196ef47696d731
|
| 3 |
size 4984
|