MHGanainy/gpt2-xl-lora-ecthr-random-imbalanced-skewed-cluster-8-id-5-blah
Browse files- README.md +2 -0
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +490 -0
README.md
CHANGED
|
@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
# gpt2-xl-lora-ecthr-random-imbalanced-skewed-cluster-8-id-5-blah
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [openai-community/gpt2-xl](https://huggingface.co/openai-community/gpt2-xl) on an unknown dataset.
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
|
|
| 15 |
# gpt2-xl-lora-ecthr-random-imbalanced-skewed-cluster-8-id-5-blah
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [openai-community/gpt2-xl](https://huggingface.co/openai-community/gpt2-xl) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 1.9495
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 1.9494917392730713,
|
| 4 |
+
"eval_runtime": 54.4179,
|
| 5 |
+
"eval_samples_per_second": 22.125,
|
| 6 |
+
"eval_steps_per_second": 2.775,
|
| 7 |
+
"perplexity": 7.025116082433238,
|
| 8 |
+
"total_flos": 1.17399030595584e+17,
|
| 9 |
+
"train_loss": 2.1091008744372672,
|
| 10 |
+
"train_runtime": 1734.4968,
|
| 11 |
+
"train_samples_per_second": 7.449,
|
| 12 |
+
"train_steps_per_second": 3.724
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 1.9494917392730713,
|
| 4 |
+
"eval_runtime": 54.4179,
|
| 5 |
+
"eval_samples_per_second": 22.125,
|
| 6 |
+
"eval_steps_per_second": 2.775,
|
| 7 |
+
"perplexity": 7.025116082433238
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.17399030595584e+17,
|
| 4 |
+
"train_loss": 2.1091008744372672,
|
| 5 |
+
"train_runtime": 1734.4968,
|
| 6 |
+
"train_samples_per_second": 7.449,
|
| 7 |
+
"train_steps_per_second": 3.724
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,490 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 6460,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.015479876160990712,
|
| 13 |
+
"grad_norm": 0.07901544123888016,
|
| 14 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 15 |
+
"loss": 2.4612,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.030959752321981424,
|
| 20 |
+
"grad_norm": 0.1504211276769638,
|
| 21 |
+
"learning_rate": 4.000000000000001e-06,
|
| 22 |
+
"loss": 2.4539,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.04643962848297214,
|
| 27 |
+
"grad_norm": 0.18693134188652039,
|
| 28 |
+
"learning_rate": 6e-06,
|
| 29 |
+
"loss": 2.4054,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06191950464396285,
|
| 34 |
+
"grad_norm": 0.18700702488422394,
|
| 35 |
+
"learning_rate": 8.000000000000001e-06,
|
| 36 |
+
"loss": 2.3977,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07739938080495357,
|
| 41 |
+
"grad_norm": 0.3383652865886688,
|
| 42 |
+
"learning_rate": 1e-05,
|
| 43 |
+
"loss": 2.355,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.09287925696594428,
|
| 48 |
+
"grad_norm": 0.3366035223007202,
|
| 49 |
+
"learning_rate": 1.2e-05,
|
| 50 |
+
"loss": 2.3473,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.10835913312693499,
|
| 55 |
+
"grad_norm": 0.38318321108818054,
|
| 56 |
+
"learning_rate": 1.4e-05,
|
| 57 |
+
"loss": 2.2959,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1238390092879257,
|
| 62 |
+
"grad_norm": 0.42260441184043884,
|
| 63 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 64 |
+
"loss": 2.2843,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1393188854489164,
|
| 69 |
+
"grad_norm": 0.44113874435424805,
|
| 70 |
+
"learning_rate": 1.8e-05,
|
| 71 |
+
"loss": 2.2534,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.15479876160990713,
|
| 76 |
+
"grad_norm": 0.7373621463775635,
|
| 77 |
+
"learning_rate": 2e-05,
|
| 78 |
+
"loss": 2.2357,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.17027863777089783,
|
| 83 |
+
"grad_norm": 0.7207284569740295,
|
| 84 |
+
"learning_rate": 1.998345127099248e-05,
|
| 85 |
+
"loss": 2.1977,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.18575851393188855,
|
| 90 |
+
"grad_norm": 0.6718271970748901,
|
| 91 |
+
"learning_rate": 1.9933859856056264e-05,
|
| 92 |
+
"loss": 2.1542,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.20123839009287925,
|
| 97 |
+
"grad_norm": 0.5868344306945801,
|
| 98 |
+
"learning_rate": 1.9851389890168738e-05,
|
| 99 |
+
"loss": 2.1822,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.21671826625386997,
|
| 104 |
+
"grad_norm": 0.6902896165847778,
|
| 105 |
+
"learning_rate": 1.9736314327953242e-05,
|
| 106 |
+
"loss": 2.177,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.23219814241486067,
|
| 111 |
+
"grad_norm": 0.5984227657318115,
|
| 112 |
+
"learning_rate": 1.9589014040268678e-05,
|
| 113 |
+
"loss": 2.1142,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2476780185758514,
|
| 118 |
+
"grad_norm": 0.7339717149734497,
|
| 119 |
+
"learning_rate": 1.9409976553623767e-05,
|
| 120 |
+
"loss": 2.16,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2631578947368421,
|
| 125 |
+
"grad_norm": 0.8267336487770081,
|
| 126 |
+
"learning_rate": 1.9199794436588244e-05,
|
| 127 |
+
"loss": 2.1297,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.2786377708978328,
|
| 132 |
+
"grad_norm": 0.9779782891273499,
|
| 133 |
+
"learning_rate": 1.895916333854152e-05,
|
| 134 |
+
"loss": 2.105,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.29411764705882354,
|
| 139 |
+
"grad_norm": 0.8321210145950317,
|
| 140 |
+
"learning_rate": 1.8688879687250067e-05,
|
| 141 |
+
"loss": 2.1347,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.30959752321981426,
|
| 146 |
+
"grad_norm": 0.6403323411941528,
|
| 147 |
+
"learning_rate": 1.838983805289396e-05,
|
| 148 |
+
"loss": 2.1528,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.32507739938080493,
|
| 153 |
+
"grad_norm": 0.9915246963500977,
|
| 154 |
+
"learning_rate": 1.8063028187266987e-05,
|
| 155 |
+
"loss": 2.097,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.34055727554179566,
|
| 160 |
+
"grad_norm": 0.8555511236190796,
|
| 161 |
+
"learning_rate": 1.7709531747949796e-05,
|
| 162 |
+
"loss": 2.1227,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.3560371517027864,
|
| 167 |
+
"grad_norm": 1.1070411205291748,
|
| 168 |
+
"learning_rate": 1.7330518718298263e-05,
|
| 169 |
+
"loss": 2.0825,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3715170278637771,
|
| 174 |
+
"grad_norm": 0.7396721839904785,
|
| 175 |
+
"learning_rate": 1.6927243535095995e-05,
|
| 176 |
+
"loss": 2.0856,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.38699690402476783,
|
| 181 |
+
"grad_norm": 0.73526930809021,
|
| 182 |
+
"learning_rate": 1.6501040936687444e-05,
|
| 183 |
+
"loss": 2.0817,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.4024767801857585,
|
| 188 |
+
"grad_norm": 1.1368354558944702,
|
| 189 |
+
"learning_rate": 1.605332154533328e-05,
|
| 190 |
+
"loss": 2.081,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.4179566563467492,
|
| 195 |
+
"grad_norm": 0.8981988430023193,
|
| 196 |
+
"learning_rate": 1.55855671984093e-05,
|
| 197 |
+
"loss": 2.0854,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.43343653250773995,
|
| 202 |
+
"grad_norm": 0.7327935099601746,
|
| 203 |
+
"learning_rate": 1.5099326043901361e-05,
|
| 204 |
+
"loss": 2.1008,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.44891640866873067,
|
| 209 |
+
"grad_norm": 0.9283792972564697,
|
| 210 |
+
"learning_rate": 1.4596207416429119e-05,
|
| 211 |
+
"loss": 2.0793,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.46439628482972134,
|
| 216 |
+
"grad_norm": 0.9368755221366882,
|
| 217 |
+
"learning_rate": 1.4077876510757502e-05,
|
| 218 |
+
"loss": 2.0227,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.47987616099071206,
|
| 223 |
+
"grad_norm": 0.9884905815124512,
|
| 224 |
+
"learning_rate": 1.3546048870425356e-05,
|
| 225 |
+
"loss": 2.0503,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.4953560371517028,
|
| 230 |
+
"grad_norm": 0.9788491129875183,
|
| 231 |
+
"learning_rate": 1.300248470973239e-05,
|
| 232 |
+
"loss": 2.0337,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.5108359133126935,
|
| 237 |
+
"grad_norm": 0.9716861844062805,
|
| 238 |
+
"learning_rate": 1.244898308787731e-05,
|
| 239 |
+
"loss": 2.0707,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.5263157894736842,
|
| 244 |
+
"grad_norm": 0.944068968296051,
|
| 245 |
+
"learning_rate": 1.1887375954529167e-05,
|
| 246 |
+
"loss": 2.1104,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.541795665634675,
|
| 251 |
+
"grad_norm": 1.0999778509140015,
|
| 252 |
+
"learning_rate": 1.1319522086539666e-05,
|
| 253 |
+
"loss": 2.0316,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.5572755417956656,
|
| 258 |
+
"grad_norm": 2.304236650466919,
|
| 259 |
+
"learning_rate": 1.0747300935864245e-05,
|
| 260 |
+
"loss": 2.0456,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.5727554179566563,
|
| 265 |
+
"grad_norm": 2.055583953857422,
|
| 266 |
+
"learning_rate": 1.0172606409053887e-05,
|
| 267 |
+
"loss": 2.0065,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.5882352941176471,
|
| 272 |
+
"grad_norm": 0.9950677156448364,
|
| 273 |
+
"learning_rate": 9.597340598905851e-06,
|
| 274 |
+
"loss": 2.026,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.6037151702786377,
|
| 279 |
+
"grad_norm": 0.9505623579025269,
|
| 280 |
+
"learning_rate": 9.02340748902002e-06,
|
| 281 |
+
"loss": 1.9838,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.6191950464396285,
|
| 286 |
+
"grad_norm": 0.6892825365066528,
|
| 287 |
+
"learning_rate": 8.452706652097187e-06,
|
| 288 |
+
"loss": 2.0418,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.6346749226006192,
|
| 293 |
+
"grad_norm": 1.3457064628601074,
|
| 294 |
+
"learning_rate": 7.887126962836274e-06,
|
| 295 |
+
"loss": 2.0684,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.6501547987616099,
|
| 300 |
+
"grad_norm": 0.9981542229652405,
|
| 301 |
+
"learning_rate": 7.328540346239222e-06,
|
| 302 |
+
"loss": 2.0349,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.6656346749226006,
|
| 307 |
+
"grad_norm": 1.0341832637786865,
|
| 308 |
+
"learning_rate": 6.778795582015096e-06,
|
| 309 |
+
"loss": 2.0187,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.6811145510835913,
|
| 314 |
+
"grad_norm": 1.1213501691818237,
|
| 315 |
+
"learning_rate": 6.239712185589181e-06,
|
| 316 |
+
"loss": 1.9938,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.6965944272445821,
|
| 321 |
+
"grad_norm": 1.9015402793884277,
|
| 322 |
+
"learning_rate": 5.713074385969457e-06,
|
| 323 |
+
"loss": 2.0723,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.7120743034055728,
|
| 328 |
+
"grad_norm": 1.2015235424041748,
|
| 329 |
+
"learning_rate": 5.200625220402139e-06,
|
| 330 |
+
"loss": 2.0688,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.7275541795665634,
|
| 335 |
+
"grad_norm": 1.0588676929473877,
|
| 336 |
+
"learning_rate": 4.7040607653614335e-06,
|
| 337 |
+
"loss": 2.0144,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.7430340557275542,
|
| 342 |
+
"grad_norm": 0.8498965501785278,
|
| 343 |
+
"learning_rate": 4.225024522967602e-06,
|
| 344 |
+
"loss": 2.0868,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.7585139318885449,
|
| 349 |
+
"grad_norm": 0.9379357695579529,
|
| 350 |
+
"learning_rate": 3.7651019814126656e-06,
|
| 351 |
+
"loss": 2.0198,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.7739938080495357,
|
| 356 |
+
"grad_norm": 1.0378938913345337,
|
| 357 |
+
"learning_rate": 3.325815367397557e-06,
|
| 358 |
+
"loss": 1.9839,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.7894736842105263,
|
| 363 |
+
"grad_norm": 0.9907466173171997,
|
| 364 |
+
"learning_rate": 2.908618607948669e-06,
|
| 365 |
+
"loss": 2.0121,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.804953560371517,
|
| 370 |
+
"grad_norm": 1.0447109937667847,
|
| 371 |
+
"learning_rate": 2.514892518288988e-06,
|
| 372 |
+
"loss": 2.0135,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.8204334365325078,
|
| 377 |
+
"grad_norm": 0.9206366539001465,
|
| 378 |
+
"learning_rate": 2.145940231690713e-06,
|
| 379 |
+
"loss": 2.0522,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.8359133126934984,
|
| 384 |
+
"grad_norm": 0.9790246486663818,
|
| 385 |
+
"learning_rate": 1.8029828864353583e-06,
|
| 386 |
+
"loss": 2.0227,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.8513931888544891,
|
| 391 |
+
"grad_norm": 1.1046849489212036,
|
| 392 |
+
"learning_rate": 1.4871555841564889e-06,
|
| 393 |
+
"loss": 2.0442,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.8668730650154799,
|
| 398 |
+
"grad_norm": 0.9425301551818848,
|
| 399 |
+
"learning_rate": 1.1995036329418153e-06,
|
| 400 |
+
"loss": 2.0347,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.8823529411764706,
|
| 405 |
+
"grad_norm": 1.1868618726730347,
|
| 406 |
+
"learning_rate": 9.409790876291658e-07,
|
| 407 |
+
"loss": 2.0145,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.8978328173374613,
|
| 412 |
+
"grad_norm": 1.109472393989563,
|
| 413 |
+
"learning_rate": 7.124375987469767e-07,
|
| 414 |
+
"loss": 1.999,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.913312693498452,
|
| 419 |
+
"grad_norm": 0.9162923097610474,
|
| 420 |
+
"learning_rate": 5.146355805285452e-07,
|
| 421 |
+
"loss": 2.0472,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.9287925696594427,
|
| 426 |
+
"grad_norm": 1.168561339378357,
|
| 427 |
+
"learning_rate": 3.482277073731988e-07,
|
| 428 |
+
"loss": 2.0582,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.9442724458204335,
|
| 433 |
+
"grad_norm": 0.9131506681442261,
|
| 434 |
+
"learning_rate": 2.1376474704044693e-07,
|
| 435 |
+
"loss": 2.0018,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.9597523219814241,
|
| 440 |
+
"grad_norm": 0.9595931172370911,
|
| 441 |
+
"learning_rate": 1.1169173774871478e-07,
|
| 442 |
+
"loss": 2.0378,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.9752321981424149,
|
| 447 |
+
"grad_norm": 0.7124084234237671,
|
| 448 |
+
"learning_rate": 4.234651521194844e-08,
|
| 449 |
+
"loss": 2.0062,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.9907120743034056,
|
| 454 |
+
"grad_norm": 1.1125590801239014,
|
| 455 |
+
"learning_rate": 5.958594489295921e-09,
|
| 456 |
+
"loss": 2.0932,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.0,
|
| 461 |
+
"step": 6460,
|
| 462 |
+
"total_flos": 1.17399030595584e+17,
|
| 463 |
+
"train_loss": 2.1091008744372672,
|
| 464 |
+
"train_runtime": 1734.4968,
|
| 465 |
+
"train_samples_per_second": 7.449,
|
| 466 |
+
"train_steps_per_second": 3.724
|
| 467 |
+
}
|
| 468 |
+
],
|
| 469 |
+
"logging_steps": 100,
|
| 470 |
+
"max_steps": 6460,
|
| 471 |
+
"num_input_tokens_seen": 0,
|
| 472 |
+
"num_train_epochs": 1,
|
| 473 |
+
"save_steps": 500,
|
| 474 |
+
"stateful_callbacks": {
|
| 475 |
+
"TrainerControl": {
|
| 476 |
+
"args": {
|
| 477 |
+
"should_epoch_stop": false,
|
| 478 |
+
"should_evaluate": false,
|
| 479 |
+
"should_log": false,
|
| 480 |
+
"should_save": true,
|
| 481 |
+
"should_training_stop": true
|
| 482 |
+
},
|
| 483 |
+
"attributes": {}
|
| 484 |
+
}
|
| 485 |
+
},
|
| 486 |
+
"total_flos": 1.17399030595584e+17,
|
| 487 |
+
"train_batch_size": 2,
|
| 488 |
+
"trial_name": null,
|
| 489 |
+
"trial_params": null
|
| 490 |
+
}
|