Update multitask INDRA statement classifier
Browse files- all_results.json +17 -17
- config.json +1 -1
- eval_results.json +17 -17
- indra_label2id.json +23 -8
- model.safetensors +2 -2
- test_results.json +17 -17
- trainer_state.json +267 -267
- training_args.bin +1 -1
all_results.json
CHANGED
|
@@ -1,22 +1,22 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_gate1_accuracy": 0.
|
| 4 |
-
"eval_gate1_f1": 0.
|
| 5 |
-
"eval_gate1_precision": 0.
|
| 6 |
-
"eval_gate1_recall": 0.
|
| 7 |
"eval_gate1_support": 5481,
|
| 8 |
-
"eval_gate2_accuracy": 0.
|
| 9 |
-
"eval_gate2_f1": 0.
|
| 10 |
-
"eval_gate2_precision": 0.
|
| 11 |
-
"eval_gate2_recall": 0.
|
| 12 |
"eval_gate2_support": 5481,
|
| 13 |
-
"eval_gate3_accuracy": 0.
|
| 14 |
-
"eval_gate3_f1": 0.
|
| 15 |
-
"eval_gate3_precision": 0.
|
| 16 |
-
"eval_gate3_recall": 0.
|
| 17 |
-
"eval_gate3_support":
|
| 18 |
-
"eval_loss": 0.
|
| 19 |
-
"eval_runtime":
|
| 20 |
-
"eval_samples_per_second":
|
| 21 |
-
"eval_steps_per_second":
|
| 22 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_gate1_accuracy": 0.9204524721766101,
|
| 4 |
+
"eval_gate1_f1": 0.9175696677724912,
|
| 5 |
+
"eval_gate1_precision": 0.9198718411071631,
|
| 6 |
+
"eval_gate1_recall": 0.9156173695177233,
|
| 7 |
"eval_gate1_support": 5481,
|
| 8 |
+
"eval_gate2_accuracy": 0.8573253056011677,
|
| 9 |
+
"eval_gate2_f1": 0.7662196765345953,
|
| 10 |
+
"eval_gate2_precision": 0.7540255243234163,
|
| 11 |
+
"eval_gate2_recall": 0.7899899686908007,
|
| 12 |
"eval_gate2_support": 5481,
|
| 13 |
+
"eval_gate3_accuracy": 0.8761520737327189,
|
| 14 |
+
"eval_gate3_f1": 0.7204269422018509,
|
| 15 |
+
"eval_gate3_precision": 0.7065840948237819,
|
| 16 |
+
"eval_gate3_recall": 0.7411799394675562,
|
| 17 |
+
"eval_gate3_support": 6944,
|
| 18 |
+
"eval_loss": 0.7167699933052063,
|
| 19 |
+
"eval_runtime": 9.951,
|
| 20 |
+
"eval_samples_per_second": 1248.614,
|
| 21 |
+
"eval_steps_per_second": 39.091
|
| 22 |
}
|
config.json
CHANGED
|
@@ -9,7 +9,7 @@
|
|
| 9 |
"gate2_loss_weight": 0.5,
|
| 10 |
"gate2_num_labels": 20,
|
| 11 |
"gate3_loss_weight": 0.25,
|
| 12 |
-
"gate3_num_labels":
|
| 13 |
"hidden_act": "gelu",
|
| 14 |
"hidden_dropout_prob": 0.1,
|
| 15 |
"hidden_size": 768,
|
|
|
|
| 9 |
"gate2_loss_weight": 0.5,
|
| 10 |
"gate2_num_labels": 20,
|
| 11 |
"gate3_loss_weight": 0.25,
|
| 12 |
+
"gate3_num_labels": 26,
|
| 13 |
"hidden_act": "gelu",
|
| 14 |
"hidden_dropout_prob": 0.1,
|
| 15 |
"hidden_size": 768,
|
eval_results.json
CHANGED
|
@@ -1,22 +1,22 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_gate1_accuracy": 0.
|
| 4 |
-
"eval_gate1_f1": 0.
|
| 5 |
-
"eval_gate1_precision": 0.
|
| 6 |
-
"eval_gate1_recall": 0.
|
| 7 |
"eval_gate1_support": 5480,
|
| 8 |
-
"eval_gate2_accuracy": 0.
|
| 9 |
-
"eval_gate2_f1": 0.
|
| 10 |
-
"eval_gate2_precision": 0.
|
| 11 |
-
"eval_gate2_recall": 0.
|
| 12 |
"eval_gate2_support": 5480,
|
| 13 |
-
"eval_gate3_accuracy": 0.
|
| 14 |
-
"eval_gate3_f1": 0.
|
| 15 |
-
"eval_gate3_precision": 0.
|
| 16 |
-
"eval_gate3_recall": 0.
|
| 17 |
-
"eval_gate3_support":
|
| 18 |
-
"eval_loss": 0.
|
| 19 |
-
"eval_runtime":
|
| 20 |
-
"eval_samples_per_second":
|
| 21 |
-
"eval_steps_per_second": 39.
|
| 22 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_gate1_accuracy": 0.9231751824817518,
|
| 4 |
+
"eval_gate1_f1": 0.91874922234221,
|
| 5 |
+
"eval_gate1_precision": 0.9213536363863121,
|
| 6 |
+
"eval_gate1_recall": 0.916446997689221,
|
| 7 |
"eval_gate1_support": 5480,
|
| 8 |
+
"eval_gate2_accuracy": 0.8593065693430657,
|
| 9 |
+
"eval_gate2_f1": 0.7763067392280264,
|
| 10 |
+
"eval_gate2_precision": 0.7921477517391428,
|
| 11 |
+
"eval_gate2_recall": 0.775442797363134,
|
| 12 |
"eval_gate2_support": 5480,
|
| 13 |
+
"eval_gate3_accuracy": 0.8810483870967742,
|
| 14 |
+
"eval_gate3_f1": 0.778218086166726,
|
| 15 |
+
"eval_gate3_precision": 0.75635112362015,
|
| 16 |
+
"eval_gate3_recall": 0.8147429373450497,
|
| 17 |
+
"eval_gate3_support": 6944,
|
| 18 |
+
"eval_loss": 0.7152989506721497,
|
| 19 |
+
"eval_runtime": 9.7809,
|
| 20 |
+
"eval_samples_per_second": 1270.237,
|
| 21 |
+
"eval_steps_per_second": 39.772
|
| 22 |
}
|
indra_label2id.json
CHANGED
|
@@ -2,12 +2,27 @@
|
|
| 2 |
"Acetylation": 0,
|
| 3 |
"Activation": 1,
|
| 4 |
"Complex": 2,
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
-
"
|
| 9 |
-
"
|
| 10 |
-
"
|
| 11 |
-
"
|
| 12 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
}
|
|
|
|
| 2 |
"Acetylation": 0,
|
| 3 |
"Activation": 1,
|
| 4 |
"Complex": 2,
|
| 5 |
+
"Deacetylation": 3,
|
| 6 |
+
"DecreaseAmount": 4,
|
| 7 |
+
"Deglycosylation": 5,
|
| 8 |
+
"Dehydroxylation": 6,
|
| 9 |
+
"Demethylation": 7,
|
| 10 |
+
"Dephosphorylation": 8,
|
| 11 |
+
"Deribosylation": 9,
|
| 12 |
+
"Desumoylation": 10,
|
| 13 |
+
"Deubiquitination": 11,
|
| 14 |
+
"Farnesylation": 12,
|
| 15 |
+
"Geranylgeranylation": 13,
|
| 16 |
+
"Glycosylation": 14,
|
| 17 |
+
"Hydroxylation": 15,
|
| 18 |
+
"IncreaseAmount": 16,
|
| 19 |
+
"Inhibition": 17,
|
| 20 |
+
"Methylation": 18,
|
| 21 |
+
"No_Relation": 19,
|
| 22 |
+
"Palmitoylation": 20,
|
| 23 |
+
"Phosphorylation": 21,
|
| 24 |
+
"Ribosylation": 22,
|
| 25 |
+
"Sumoylation": 23,
|
| 26 |
+
"Translocation": 24,
|
| 27 |
+
"Ubiquitination": 25
|
| 28 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3764aec317c8d33aa9d03c8e6c81eab6a79d179058f2c5d3b285e08b28f74e6
|
| 3 |
+
size 433108512
|
test_results.json
CHANGED
|
@@ -1,22 +1,22 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_gate1_accuracy": 0.
|
| 4 |
-
"eval_gate1_f1": 0.
|
| 5 |
-
"eval_gate1_precision": 0.
|
| 6 |
-
"eval_gate1_recall": 0.
|
| 7 |
"eval_gate1_support": 5481,
|
| 8 |
-
"eval_gate2_accuracy": 0.
|
| 9 |
-
"eval_gate2_f1": 0.
|
| 10 |
-
"eval_gate2_precision": 0.
|
| 11 |
-
"eval_gate2_recall": 0.
|
| 12 |
"eval_gate2_support": 5481,
|
| 13 |
-
"eval_gate3_accuracy": 0.
|
| 14 |
-
"eval_gate3_f1": 0.
|
| 15 |
-
"eval_gate3_precision": 0.
|
| 16 |
-
"eval_gate3_recall": 0.
|
| 17 |
-
"eval_gate3_support":
|
| 18 |
-
"eval_loss": 0.
|
| 19 |
-
"eval_runtime":
|
| 20 |
-
"eval_samples_per_second":
|
| 21 |
-
"eval_steps_per_second":
|
| 22 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_gate1_accuracy": 0.9204524721766101,
|
| 4 |
+
"eval_gate1_f1": 0.9175696677724912,
|
| 5 |
+
"eval_gate1_precision": 0.9198718411071631,
|
| 6 |
+
"eval_gate1_recall": 0.9156173695177233,
|
| 7 |
"eval_gate1_support": 5481,
|
| 8 |
+
"eval_gate2_accuracy": 0.8573253056011677,
|
| 9 |
+
"eval_gate2_f1": 0.7662196765345953,
|
| 10 |
+
"eval_gate2_precision": 0.7540255243234163,
|
| 11 |
+
"eval_gate2_recall": 0.7899899686908007,
|
| 12 |
"eval_gate2_support": 5481,
|
| 13 |
+
"eval_gate3_accuracy": 0.8761520737327189,
|
| 14 |
+
"eval_gate3_f1": 0.7204269422018509,
|
| 15 |
+
"eval_gate3_precision": 0.7065840948237819,
|
| 16 |
+
"eval_gate3_recall": 0.7411799394675562,
|
| 17 |
+
"eval_gate3_support": 6944,
|
| 18 |
+
"eval_loss": 0.7167699933052063,
|
| 19 |
+
"eval_runtime": 9.951,
|
| 20 |
+
"eval_samples_per_second": 1248.614,
|
| 21 |
+
"eval_steps_per_second": 39.091
|
| 22 |
}
|
trainer_state.json
CHANGED
|
@@ -1,372 +1,372 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 0.
|
| 4 |
-
"best_model_checkpoint": "output/indra_stmt_classifier/checkpoint-
|
| 5 |
"epoch": 10.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 1.0,
|
| 14 |
-
"grad_norm":
|
| 15 |
-
"learning_rate": 2.
|
| 16 |
-
"loss": 1.
|
| 17 |
-
"step":
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 1.0,
|
| 21 |
-
"eval_gate1_accuracy": 0.
|
| 22 |
-
"eval_gate1_f1": 0.
|
| 23 |
-
"eval_gate1_precision": 0.
|
| 24 |
-
"eval_gate1_recall": 0.
|
| 25 |
"eval_gate1_support": 5480,
|
| 26 |
-
"eval_gate2_accuracy": 0.
|
| 27 |
-
"eval_gate2_f1": 0.
|
| 28 |
-
"eval_gate2_precision": 0.
|
| 29 |
-
"eval_gate2_recall": 0.
|
| 30 |
"eval_gate2_support": 5480,
|
| 31 |
-
"eval_gate3_accuracy": 0.
|
| 32 |
-
"eval_gate3_f1": 0.
|
| 33 |
-
"eval_gate3_precision": 0.
|
| 34 |
-
"eval_gate3_recall": 0.
|
| 35 |
-
"eval_gate3_support":
|
| 36 |
-
"eval_loss": 0.
|
| 37 |
-
"eval_runtime":
|
| 38 |
-
"eval_samples_per_second":
|
| 39 |
"eval_steps_per_second": 39.684,
|
| 40 |
-
"step":
|
| 41 |
},
|
| 42 |
{
|
| 43 |
"epoch": 2.0,
|
| 44 |
-
"grad_norm":
|
| 45 |
-
"learning_rate": 2.
|
| 46 |
-
"loss": 0.
|
| 47 |
-
"step":
|
| 48 |
},
|
| 49 |
{
|
| 50 |
"epoch": 2.0,
|
| 51 |
-
"eval_gate1_accuracy": 0.
|
| 52 |
-
"eval_gate1_f1": 0.
|
| 53 |
-
"eval_gate1_precision": 0.
|
| 54 |
-
"eval_gate1_recall": 0.
|
| 55 |
"eval_gate1_support": 5480,
|
| 56 |
-
"eval_gate2_accuracy": 0.
|
| 57 |
-
"eval_gate2_f1": 0.
|
| 58 |
-
"eval_gate2_precision": 0.
|
| 59 |
-
"eval_gate2_recall": 0.
|
| 60 |
"eval_gate2_support": 5480,
|
| 61 |
-
"eval_gate3_accuracy": 0.
|
| 62 |
-
"eval_gate3_f1": 0.
|
| 63 |
-
"eval_gate3_precision": 0.
|
| 64 |
-
"eval_gate3_recall": 0.
|
| 65 |
-
"eval_gate3_support":
|
| 66 |
-
"eval_loss": 0.
|
| 67 |
-
"eval_runtime":
|
| 68 |
-
"eval_samples_per_second":
|
| 69 |
-
"eval_steps_per_second":
|
| 70 |
-
"step":
|
| 71 |
},
|
| 72 |
{
|
| 73 |
"epoch": 3.0,
|
| 74 |
-
"grad_norm":
|
| 75 |
-
"learning_rate": 2.
|
| 76 |
-
"loss": 0.
|
| 77 |
-
"step":
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 3.0,
|
| 81 |
-
"eval_gate1_accuracy": 0.
|
| 82 |
-
"eval_gate1_f1": 0.
|
| 83 |
-
"eval_gate1_precision": 0.
|
| 84 |
-
"eval_gate1_recall": 0.
|
| 85 |
"eval_gate1_support": 5480,
|
| 86 |
-
"eval_gate2_accuracy": 0.
|
| 87 |
-
"eval_gate2_f1": 0.
|
| 88 |
-
"eval_gate2_precision": 0.
|
| 89 |
-
"eval_gate2_recall": 0.
|
| 90 |
"eval_gate2_support": 5480,
|
| 91 |
-
"eval_gate3_accuracy": 0.
|
| 92 |
-
"eval_gate3_f1": 0.
|
| 93 |
-
"eval_gate3_precision": 0.
|
| 94 |
-
"eval_gate3_recall": 0.
|
| 95 |
-
"eval_gate3_support":
|
| 96 |
-
"eval_loss": 0.
|
| 97 |
-
"eval_runtime":
|
| 98 |
-
"eval_samples_per_second":
|
| 99 |
-
"eval_steps_per_second":
|
| 100 |
-
"step":
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 4.0,
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate": 1.
|
| 106 |
-
"loss": 0.
|
| 107 |
-
"step":
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 4.0,
|
| 111 |
-
"eval_gate1_accuracy": 0.
|
| 112 |
-
"eval_gate1_f1": 0.
|
| 113 |
-
"eval_gate1_precision": 0.
|
| 114 |
-
"eval_gate1_recall": 0.
|
| 115 |
"eval_gate1_support": 5480,
|
| 116 |
-
"eval_gate2_accuracy": 0.
|
| 117 |
-
"eval_gate2_f1": 0.
|
| 118 |
-
"eval_gate2_precision": 0.
|
| 119 |
-
"eval_gate2_recall": 0.
|
| 120 |
"eval_gate2_support": 5480,
|
| 121 |
-
"eval_gate3_accuracy": 0.
|
| 122 |
-
"eval_gate3_f1": 0.
|
| 123 |
-
"eval_gate3_precision": 0.
|
| 124 |
-
"eval_gate3_recall": 0.
|
| 125 |
-
"eval_gate3_support":
|
| 126 |
-
"eval_loss": 0.
|
| 127 |
-
"eval_runtime":
|
| 128 |
-
"eval_samples_per_second":
|
| 129 |
-
"eval_steps_per_second":
|
| 130 |
-
"step":
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 5.0,
|
| 134 |
-
"grad_norm": 5.
|
| 135 |
-
"learning_rate": 1.
|
| 136 |
-
"loss": 0.
|
| 137 |
-
"step":
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 5.0,
|
| 141 |
-
"eval_gate1_accuracy": 0.
|
| 142 |
-
"eval_gate1_f1": 0.
|
| 143 |
-
"eval_gate1_precision": 0.
|
| 144 |
-
"eval_gate1_recall": 0.
|
| 145 |
"eval_gate1_support": 5480,
|
| 146 |
-
"eval_gate2_accuracy": 0.
|
| 147 |
-
"eval_gate2_f1": 0.
|
| 148 |
-
"eval_gate2_precision": 0.
|
| 149 |
-
"eval_gate2_recall": 0.
|
| 150 |
"eval_gate2_support": 5480,
|
| 151 |
-
"eval_gate3_accuracy": 0.
|
| 152 |
-
"eval_gate3_f1": 0.
|
| 153 |
-
"eval_gate3_precision": 0.
|
| 154 |
-
"eval_gate3_recall": 0.
|
| 155 |
-
"eval_gate3_support":
|
| 156 |
-
"eval_loss": 0.
|
| 157 |
-
"eval_runtime":
|
| 158 |
-
"eval_samples_per_second":
|
| 159 |
-
"eval_steps_per_second": 40.
|
| 160 |
-
"step":
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 6.0,
|
| 164 |
-
"grad_norm":
|
| 165 |
-
"learning_rate": 1.
|
| 166 |
-
"loss": 0.
|
| 167 |
-
"step":
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 6.0,
|
| 171 |
-
"eval_gate1_accuracy": 0.
|
| 172 |
-
"eval_gate1_f1": 0.
|
| 173 |
-
"eval_gate1_precision": 0.
|
| 174 |
-
"eval_gate1_recall": 0.
|
| 175 |
"eval_gate1_support": 5480,
|
| 176 |
-
"eval_gate2_accuracy": 0.
|
| 177 |
-
"eval_gate2_f1": 0.
|
| 178 |
-
"eval_gate2_precision": 0.
|
| 179 |
-
"eval_gate2_recall": 0.
|
| 180 |
"eval_gate2_support": 5480,
|
| 181 |
-
"eval_gate3_accuracy": 0.
|
| 182 |
-
"eval_gate3_f1": 0.
|
| 183 |
-
"eval_gate3_precision": 0.
|
| 184 |
-
"eval_gate3_recall": 0.
|
| 185 |
-
"eval_gate3_support":
|
| 186 |
-
"eval_loss": 0.
|
| 187 |
-
"eval_runtime":
|
| 188 |
-
"eval_samples_per_second":
|
| 189 |
-
"eval_steps_per_second":
|
| 190 |
-
"step":
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 7.0,
|
| 194 |
-
"grad_norm":
|
| 195 |
-
"learning_rate": 9.
|
| 196 |
-
"loss": 0.
|
| 197 |
-
"step":
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 7.0,
|
| 201 |
-
"eval_gate1_accuracy": 0.
|
| 202 |
-
"eval_gate1_f1": 0.
|
| 203 |
-
"eval_gate1_precision": 0.
|
| 204 |
-
"eval_gate1_recall": 0.
|
| 205 |
"eval_gate1_support": 5480,
|
| 206 |
-
"eval_gate2_accuracy": 0.
|
| 207 |
-
"eval_gate2_f1": 0.
|
| 208 |
-
"eval_gate2_precision": 0.
|
| 209 |
-
"eval_gate2_recall": 0.
|
| 210 |
"eval_gate2_support": 5480,
|
| 211 |
-
"eval_gate3_accuracy": 0.
|
| 212 |
-
"eval_gate3_f1": 0.
|
| 213 |
-
"eval_gate3_precision": 0.
|
| 214 |
-
"eval_gate3_recall": 0.
|
| 215 |
-
"eval_gate3_support":
|
| 216 |
-
"eval_loss": 0.
|
| 217 |
-
"eval_runtime":
|
| 218 |
-
"eval_samples_per_second":
|
| 219 |
-
"eval_steps_per_second": 39.
|
| 220 |
-
"step":
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 8.0,
|
| 224 |
-
"grad_norm":
|
| 225 |
-
"learning_rate": 6.
|
| 226 |
-
"loss": 0.
|
| 227 |
-
"step":
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 8.0,
|
| 231 |
-
"eval_gate1_accuracy": 0.
|
| 232 |
-
"eval_gate1_f1": 0.
|
| 233 |
-
"eval_gate1_precision": 0.
|
| 234 |
-
"eval_gate1_recall": 0.
|
| 235 |
"eval_gate1_support": 5480,
|
| 236 |
-
"eval_gate2_accuracy": 0.
|
| 237 |
-
"eval_gate2_f1": 0.
|
| 238 |
-
"eval_gate2_precision": 0.
|
| 239 |
-
"eval_gate2_recall": 0.
|
| 240 |
"eval_gate2_support": 5480,
|
| 241 |
-
"eval_gate3_accuracy": 0.
|
| 242 |
-
"eval_gate3_f1": 0.
|
| 243 |
-
"eval_gate3_precision": 0.
|
| 244 |
-
"eval_gate3_recall": 0.
|
| 245 |
-
"eval_gate3_support":
|
| 246 |
-
"eval_loss": 0.
|
| 247 |
-
"eval_runtime":
|
| 248 |
-
"eval_samples_per_second":
|
| 249 |
-
"eval_steps_per_second":
|
| 250 |
-
"step":
|
| 251 |
},
|
| 252 |
{
|
| 253 |
"epoch": 9.0,
|
| 254 |
-
"grad_norm":
|
| 255 |
-
"learning_rate": 3.
|
| 256 |
-
"loss": 0.
|
| 257 |
-
"step":
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 9.0,
|
| 261 |
-
"eval_gate1_accuracy": 0.
|
| 262 |
-
"eval_gate1_f1": 0.
|
| 263 |
-
"eval_gate1_precision": 0.
|
| 264 |
-
"eval_gate1_recall": 0.
|
| 265 |
"eval_gate1_support": 5480,
|
| 266 |
-
"eval_gate2_accuracy": 0.
|
| 267 |
-
"eval_gate2_f1": 0.
|
| 268 |
-
"eval_gate2_precision": 0.
|
| 269 |
-
"eval_gate2_recall": 0.
|
| 270 |
"eval_gate2_support": 5480,
|
| 271 |
-
"eval_gate3_accuracy": 0.
|
| 272 |
-
"eval_gate3_f1": 0.
|
| 273 |
-
"eval_gate3_precision": 0.
|
| 274 |
-
"eval_gate3_recall": 0.
|
| 275 |
-
"eval_gate3_support":
|
| 276 |
-
"eval_loss": 0.
|
| 277 |
-
"eval_runtime":
|
| 278 |
-
"eval_samples_per_second":
|
| 279 |
-
"eval_steps_per_second": 40.
|
| 280 |
-
"step":
|
| 281 |
},
|
| 282 |
{
|
| 283 |
"epoch": 10.0,
|
| 284 |
-
"grad_norm":
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss": 0.
|
| 287 |
-
"step":
|
| 288 |
},
|
| 289 |
{
|
| 290 |
"epoch": 10.0,
|
| 291 |
-
"eval_gate1_accuracy": 0.
|
| 292 |
-
"eval_gate1_f1": 0.
|
| 293 |
-
"eval_gate1_precision": 0.
|
| 294 |
-
"eval_gate1_recall": 0.
|
| 295 |
"eval_gate1_support": 5480,
|
| 296 |
-
"eval_gate2_accuracy": 0.
|
| 297 |
-
"eval_gate2_f1": 0.
|
| 298 |
-
"eval_gate2_precision": 0.
|
| 299 |
-
"eval_gate2_recall": 0.
|
| 300 |
"eval_gate2_support": 5480,
|
| 301 |
-
"eval_gate3_accuracy": 0.
|
| 302 |
-
"eval_gate3_f1": 0.
|
| 303 |
-
"eval_gate3_precision": 0.
|
| 304 |
-
"eval_gate3_recall": 0.
|
| 305 |
-
"eval_gate3_support":
|
| 306 |
-
"eval_loss": 1.
|
| 307 |
-
"eval_runtime":
|
| 308 |
-
"eval_samples_per_second":
|
| 309 |
-
"eval_steps_per_second":
|
| 310 |
-
"step":
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 10.0,
|
| 314 |
-
"step":
|
| 315 |
-
"total_flos":
|
| 316 |
-
"train_loss": 0.
|
| 317 |
-
"train_runtime":
|
| 318 |
-
"train_samples_per_second":
|
| 319 |
-
"train_steps_per_second": 13.
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 10.0,
|
| 323 |
-
"eval_gate1_accuracy": 0.
|
| 324 |
-
"eval_gate1_f1": 0.
|
| 325 |
-
"eval_gate1_precision": 0.
|
| 326 |
-
"eval_gate1_recall": 0.
|
| 327 |
"eval_gate1_support": 5480,
|
| 328 |
-
"eval_gate2_accuracy": 0.
|
| 329 |
-
"eval_gate2_f1": 0.
|
| 330 |
-
"eval_gate2_precision": 0.
|
| 331 |
-
"eval_gate2_recall": 0.
|
| 332 |
"eval_gate2_support": 5480,
|
| 333 |
-
"eval_gate3_accuracy": 0.
|
| 334 |
-
"eval_gate3_f1": 0.
|
| 335 |
-
"eval_gate3_precision": 0.
|
| 336 |
-
"eval_gate3_recall": 0.
|
| 337 |
-
"eval_gate3_support":
|
| 338 |
-
"eval_loss": 0.
|
| 339 |
-
"eval_runtime":
|
| 340 |
-
"eval_samples_per_second":
|
| 341 |
-
"eval_steps_per_second": 39.
|
| 342 |
-
"step":
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 10.0,
|
| 346 |
-
"eval_gate1_accuracy": 0.
|
| 347 |
-
"eval_gate1_f1": 0.
|
| 348 |
-
"eval_gate1_precision": 0.
|
| 349 |
-
"eval_gate1_recall": 0.
|
| 350 |
"eval_gate1_support": 5481,
|
| 351 |
-
"eval_gate2_accuracy": 0.
|
| 352 |
-
"eval_gate2_f1": 0.
|
| 353 |
-
"eval_gate2_precision": 0.
|
| 354 |
-
"eval_gate2_recall": 0.
|
| 355 |
"eval_gate2_support": 5481,
|
| 356 |
-
"eval_gate3_accuracy": 0.
|
| 357 |
-
"eval_gate3_f1": 0.
|
| 358 |
-
"eval_gate3_precision": 0.
|
| 359 |
-
"eval_gate3_recall": 0.
|
| 360 |
-
"eval_gate3_support":
|
| 361 |
-
"eval_loss": 0.
|
| 362 |
-
"eval_runtime":
|
| 363 |
-
"eval_samples_per_second":
|
| 364 |
-
"eval_steps_per_second":
|
| 365 |
-
"step":
|
| 366 |
}
|
| 367 |
],
|
| 368 |
"logging_steps": 500,
|
| 369 |
-
"max_steps":
|
| 370 |
"num_input_tokens_seen": 0,
|
| 371 |
"num_train_epochs": 10,
|
| 372 |
"save_steps": 500,
|
|
@@ -382,7 +382,7 @@
|
|
| 382 |
"attributes": {}
|
| 383 |
}
|
| 384 |
},
|
| 385 |
-
"total_flos":
|
| 386 |
"train_batch_size": 32,
|
| 387 |
"trial_name": null,
|
| 388 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 15535,
|
| 3 |
+
"best_metric": 0.91874922234221,
|
| 4 |
+
"best_model_checkpoint": "output/indra_stmt_classifier/checkpoint-15535",
|
| 5 |
"epoch": 10.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 31070,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 1.0,
|
| 14 |
+
"grad_norm": 2.75587797164917,
|
| 15 |
+
"learning_rate": 2.7000965561635017e-05,
|
| 16 |
+
"loss": 1.1186,
|
| 17 |
+
"step": 3107
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 1.0,
|
| 21 |
+
"eval_gate1_accuracy": 0.8848540145985402,
|
| 22 |
+
"eval_gate1_f1": 0.8785447299140416,
|
| 23 |
+
"eval_gate1_precision": 0.879748756806488,
|
| 24 |
+
"eval_gate1_recall": 0.8774234499312313,
|
| 25 |
"eval_gate1_support": 5480,
|
| 26 |
+
"eval_gate2_accuracy": 0.7870437956204379,
|
| 27 |
+
"eval_gate2_f1": 0.5722188113150324,
|
| 28 |
+
"eval_gate2_precision": 0.5654852652404123,
|
| 29 |
+
"eval_gate2_recall": 0.5860432299031674,
|
| 30 |
"eval_gate2_support": 5480,
|
| 31 |
+
"eval_gate3_accuracy": 0.8329493087557603,
|
| 32 |
+
"eval_gate3_f1": 0.639519347317039,
|
| 33 |
+
"eval_gate3_precision": 0.677673907531002,
|
| 34 |
+
"eval_gate3_recall": 0.6297629647896671,
|
| 35 |
+
"eval_gate3_support": 6944,
|
| 36 |
+
"eval_loss": 0.7573269009590149,
|
| 37 |
+
"eval_runtime": 9.8025,
|
| 38 |
+
"eval_samples_per_second": 1267.429,
|
| 39 |
"eval_steps_per_second": 39.684,
|
| 40 |
+
"step": 3107
|
| 41 |
},
|
| 42 |
{
|
| 43 |
"epoch": 2.0,
|
| 44 |
+
"grad_norm": 9.97591781616211,
|
| 45 |
+
"learning_rate": 2.4000965561635018e-05,
|
| 46 |
+
"loss": 0.6341,
|
| 47 |
+
"step": 6214
|
| 48 |
},
|
| 49 |
{
|
| 50 |
"epoch": 2.0,
|
| 51 |
+
"eval_gate1_accuracy": 0.8994525547445256,
|
| 52 |
+
"eval_gate1_f1": 0.8920945233221212,
|
| 53 |
+
"eval_gate1_precision": 0.9025475703450652,
|
| 54 |
+
"eval_gate1_recall": 0.8850382209313125,
|
| 55 |
"eval_gate1_support": 5480,
|
| 56 |
+
"eval_gate2_accuracy": 0.8173357664233577,
|
| 57 |
+
"eval_gate2_f1": 0.6503078056225045,
|
| 58 |
+
"eval_gate2_precision": 0.6679383283455959,
|
| 59 |
+
"eval_gate2_recall": 0.6597425206224358,
|
| 60 |
"eval_gate2_support": 5480,
|
| 61 |
+
"eval_gate3_accuracy": 0.8425979262672811,
|
| 62 |
+
"eval_gate3_f1": 0.7516576653132845,
|
| 63 |
+
"eval_gate3_precision": 0.7289310992313269,
|
| 64 |
+
"eval_gate3_recall": 0.7877551283882368,
|
| 65 |
+
"eval_gate3_support": 6944,
|
| 66 |
+
"eval_loss": 0.681019127368927,
|
| 67 |
+
"eval_runtime": 9.815,
|
| 68 |
+
"eval_samples_per_second": 1265.816,
|
| 69 |
+
"eval_steps_per_second": 39.633,
|
| 70 |
+
"step": 6214
|
| 71 |
},
|
| 72 |
{
|
| 73 |
"epoch": 3.0,
|
| 74 |
+
"grad_norm": 6.9430036544799805,
|
| 75 |
+
"learning_rate": 2.100096556163502e-05,
|
| 76 |
+
"loss": 0.4397,
|
| 77 |
+
"step": 9321
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 3.0,
|
| 81 |
+
"eval_gate1_accuracy": 0.9094890510948905,
|
| 82 |
+
"eval_gate1_f1": 0.9044200252650033,
|
| 83 |
+
"eval_gate1_precision": 0.9062240634591001,
|
| 84 |
+
"eval_gate1_recall": 0.90277723499615,
|
| 85 |
"eval_gate1_support": 5480,
|
| 86 |
+
"eval_gate2_accuracy": 0.839051094890511,
|
| 87 |
+
"eval_gate2_f1": 0.7263650484208171,
|
| 88 |
+
"eval_gate2_precision": 0.7759514146848825,
|
| 89 |
+
"eval_gate2_recall": 0.7203927286117487,
|
| 90 |
"eval_gate2_support": 5480,
|
| 91 |
+
"eval_gate3_accuracy": 0.8652073732718893,
|
| 92 |
+
"eval_gate3_f1": 0.7603754676303877,
|
| 93 |
+
"eval_gate3_precision": 0.733971679937432,
|
| 94 |
+
"eval_gate3_recall": 0.8035307301942258,
|
| 95 |
+
"eval_gate3_support": 6944,
|
| 96 |
+
"eval_loss": 0.6797090172767639,
|
| 97 |
+
"eval_runtime": 9.843,
|
| 98 |
+
"eval_samples_per_second": 1262.211,
|
| 99 |
+
"eval_steps_per_second": 39.52,
|
| 100 |
+
"step": 9321
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 4.0,
|
| 104 |
+
"grad_norm": 2.871424913406372,
|
| 105 |
+
"learning_rate": 1.8000965561635018e-05,
|
| 106 |
+
"loss": 0.3263,
|
| 107 |
+
"step": 12428
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 4.0,
|
| 111 |
+
"eval_gate1_accuracy": 0.9138686131386862,
|
| 112 |
+
"eval_gate1_f1": 0.9095152677593289,
|
| 113 |
+
"eval_gate1_precision": 0.9091069849690538,
|
| 114 |
+
"eval_gate1_recall": 0.909933548120778,
|
| 115 |
"eval_gate1_support": 5480,
|
| 116 |
+
"eval_gate2_accuracy": 0.8523722627737226,
|
| 117 |
+
"eval_gate2_f1": 0.7611458139063093,
|
| 118 |
+
"eval_gate2_precision": 0.8024369787744103,
|
| 119 |
+
"eval_gate2_recall": 0.7429072801122445,
|
| 120 |
"eval_gate2_support": 5480,
|
| 121 |
+
"eval_gate3_accuracy": 0.8620391705069125,
|
| 122 |
+
"eval_gate3_f1": 0.7744302908340429,
|
| 123 |
+
"eval_gate3_precision": 0.7519119769092221,
|
| 124 |
+
"eval_gate3_recall": 0.811903740069826,
|
| 125 |
+
"eval_gate3_support": 6944,
|
| 126 |
+
"eval_loss": 0.6769687533378601,
|
| 127 |
+
"eval_runtime": 9.7569,
|
| 128 |
+
"eval_samples_per_second": 1273.351,
|
| 129 |
+
"eval_steps_per_second": 39.869,
|
| 130 |
+
"step": 12428
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 5.0,
|
| 134 |
+
"grad_norm": 5.60059928894043,
|
| 135 |
+
"learning_rate": 1.5000965561635018e-05,
|
| 136 |
+
"loss": 0.2441,
|
| 137 |
+
"step": 15535
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 5.0,
|
| 141 |
+
"eval_gate1_accuracy": 0.9231751824817518,
|
| 142 |
+
"eval_gate1_f1": 0.91874922234221,
|
| 143 |
+
"eval_gate1_precision": 0.9213536363863121,
|
| 144 |
+
"eval_gate1_recall": 0.916446997689221,
|
| 145 |
"eval_gate1_support": 5480,
|
| 146 |
+
"eval_gate2_accuracy": 0.8593065693430657,
|
| 147 |
+
"eval_gate2_f1": 0.7763067392280264,
|
| 148 |
+
"eval_gate2_precision": 0.7921477517391428,
|
| 149 |
+
"eval_gate2_recall": 0.775442797363134,
|
| 150 |
"eval_gate2_support": 5480,
|
| 151 |
+
"eval_gate3_accuracy": 0.8810483870967742,
|
| 152 |
+
"eval_gate3_f1": 0.778218086166726,
|
| 153 |
+
"eval_gate3_precision": 0.75635112362015,
|
| 154 |
+
"eval_gate3_recall": 0.8147429373450497,
|
| 155 |
+
"eval_gate3_support": 6944,
|
| 156 |
+
"eval_loss": 0.7152989506721497,
|
| 157 |
+
"eval_runtime": 9.6947,
|
| 158 |
+
"eval_samples_per_second": 1281.524,
|
| 159 |
+
"eval_steps_per_second": 40.125,
|
| 160 |
+
"step": 15535
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 6.0,
|
| 164 |
+
"grad_norm": 16.034574508666992,
|
| 165 |
+
"learning_rate": 1.2000965561635018e-05,
|
| 166 |
+
"loss": 0.1855,
|
| 167 |
+
"step": 18642
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 6.0,
|
| 171 |
+
"eval_gate1_accuracy": 0.920985401459854,
|
| 172 |
+
"eval_gate1_f1": 0.9165971625446729,
|
| 173 |
+
"eval_gate1_precision": 0.9182460374588926,
|
| 174 |
+
"eval_gate1_recall": 0.9150794681328354,
|
| 175 |
"eval_gate1_support": 5480,
|
| 176 |
+
"eval_gate2_accuracy": 0.862956204379562,
|
| 177 |
+
"eval_gate2_f1": 0.7924841981022327,
|
| 178 |
+
"eval_gate2_precision": 0.8068279196629584,
|
| 179 |
+
"eval_gate2_recall": 0.7874456055649913,
|
| 180 |
"eval_gate2_support": 5480,
|
| 181 |
+
"eval_gate3_accuracy": 0.8882488479262672,
|
| 182 |
+
"eval_gate3_f1": 0.7926910665874005,
|
| 183 |
+
"eval_gate3_precision": 0.7755825529327147,
|
| 184 |
+
"eval_gate3_recall": 0.8216291424595251,
|
| 185 |
+
"eval_gate3_support": 6944,
|
| 186 |
+
"eval_loss": 0.8007788062095642,
|
| 187 |
+
"eval_runtime": 9.803,
|
| 188 |
+
"eval_samples_per_second": 1267.368,
|
| 189 |
+
"eval_steps_per_second": 39.682,
|
| 190 |
+
"step": 18642
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 7.0,
|
| 194 |
+
"grad_norm": 11.89713191986084,
|
| 195 |
+
"learning_rate": 9.000965561635018e-06,
|
| 196 |
+
"loss": 0.1454,
|
| 197 |
+
"step": 21749
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 7.0,
|
| 201 |
+
"eval_gate1_accuracy": 0.9193430656934306,
|
| 202 |
+
"eval_gate1_f1": 0.9147193543716858,
|
| 203 |
+
"eval_gate1_precision": 0.9171674547406892,
|
| 204 |
+
"eval_gate1_recall": 0.9125446429134221,
|
| 205 |
"eval_gate1_support": 5480,
|
| 206 |
+
"eval_gate2_accuracy": 0.8640510948905109,
|
| 207 |
+
"eval_gate2_f1": 0.7946506784893311,
|
| 208 |
+
"eval_gate2_precision": 0.8100667075746566,
|
| 209 |
+
"eval_gate2_recall": 0.7852244900603023,
|
| 210 |
"eval_gate2_support": 5480,
|
| 211 |
+
"eval_gate3_accuracy": 0.8895449308755761,
|
| 212 |
+
"eval_gate3_f1": 0.7920894399575168,
|
| 213 |
+
"eval_gate3_precision": 0.7849169985769698,
|
| 214 |
+
"eval_gate3_recall": 0.8118891174634335,
|
| 215 |
+
"eval_gate3_support": 6944,
|
| 216 |
+
"eval_loss": 0.8615767955780029,
|
| 217 |
+
"eval_runtime": 9.8147,
|
| 218 |
+
"eval_samples_per_second": 1265.852,
|
| 219 |
+
"eval_steps_per_second": 39.634,
|
| 220 |
+
"step": 21749
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 8.0,
|
| 224 |
+
"grad_norm": 0.0030767242424190044,
|
| 225 |
+
"learning_rate": 6.000965561635018e-06,
|
| 226 |
+
"loss": 0.1081,
|
| 227 |
+
"step": 24856
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 8.0,
|
| 231 |
+
"eval_gate1_accuracy": 0.9198905109489051,
|
| 232 |
+
"eval_gate1_f1": 0.9149957554765835,
|
| 233 |
+
"eval_gate1_precision": 0.9192622220801286,
|
| 234 |
+
"eval_gate1_recall": 0.9114623713660069,
|
| 235 |
"eval_gate1_support": 5480,
|
| 236 |
+
"eval_gate2_accuracy": 0.8638686131386861,
|
| 237 |
+
"eval_gate2_f1": 0.7968208804038338,
|
| 238 |
+
"eval_gate2_precision": 0.7930637027689245,
|
| 239 |
+
"eval_gate2_recall": 0.8065230155153877,
|
| 240 |
"eval_gate2_support": 5480,
|
| 241 |
+
"eval_gate3_accuracy": 0.8922811059907834,
|
| 242 |
+
"eval_gate3_f1": 0.8026929577988032,
|
| 243 |
+
"eval_gate3_precision": 0.7897651226837507,
|
| 244 |
+
"eval_gate3_recall": 0.8257808045726754,
|
| 245 |
+
"eval_gate3_support": 6944,
|
| 246 |
+
"eval_loss": 0.9336212873458862,
|
| 247 |
+
"eval_runtime": 9.7984,
|
| 248 |
+
"eval_samples_per_second": 1267.963,
|
| 249 |
+
"eval_steps_per_second": 39.7,
|
| 250 |
+
"step": 24856
|
| 251 |
},
|
| 252 |
{
|
| 253 |
"epoch": 9.0,
|
| 254 |
+
"grad_norm": 116.7902603149414,
|
| 255 |
+
"learning_rate": 3.0009655616350177e-06,
|
| 256 |
+
"loss": 0.0842,
|
| 257 |
+
"step": 27963
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 9.0,
|
| 261 |
+
"eval_gate1_accuracy": 0.9215328467153284,
|
| 262 |
+
"eval_gate1_f1": 0.9169295960347313,
|
| 263 |
+
"eval_gate1_precision": 0.9200202922521744,
|
| 264 |
+
"eval_gate1_recall": 0.9142522689322582,
|
| 265 |
"eval_gate1_support": 5480,
|
| 266 |
+
"eval_gate2_accuracy": 0.8656934306569343,
|
| 267 |
+
"eval_gate2_f1": 0.7916047345282854,
|
| 268 |
+
"eval_gate2_precision": 0.7975761529206665,
|
| 269 |
+
"eval_gate2_recall": 0.7905660788693317,
|
| 270 |
"eval_gate2_support": 5480,
|
| 271 |
+
"eval_gate3_accuracy": 0.8908410138248848,
|
| 272 |
+
"eval_gate3_f1": 0.7962561702350882,
|
| 273 |
+
"eval_gate3_precision": 0.7845233050024135,
|
| 274 |
+
"eval_gate3_recall": 0.8178146646821465,
|
| 275 |
+
"eval_gate3_support": 6944,
|
| 276 |
+
"eval_loss": 0.9843125939369202,
|
| 277 |
+
"eval_runtime": 9.6312,
|
| 278 |
+
"eval_samples_per_second": 1289.977,
|
| 279 |
+
"eval_steps_per_second": 40.39,
|
| 280 |
+
"step": 27963
|
| 281 |
},
|
| 282 |
{
|
| 283 |
"epoch": 10.0,
|
| 284 |
+
"grad_norm": 0.09423983097076416,
|
| 285 |
+
"learning_rate": 9.655616350177021e-10,
|
| 286 |
+
"loss": 0.068,
|
| 287 |
+
"step": 31070
|
| 288 |
},
|
| 289 |
{
|
| 290 |
"epoch": 10.0,
|
| 291 |
+
"eval_gate1_accuracy": 0.920985401459854,
|
| 292 |
+
"eval_gate1_f1": 0.916357639651535,
|
| 293 |
+
"eval_gate1_precision": 0.9193983763664938,
|
| 294 |
+
"eval_gate1_recall": 0.9137190822830333,
|
| 295 |
"eval_gate1_support": 5480,
|
| 296 |
+
"eval_gate2_accuracy": 0.8633211678832117,
|
| 297 |
+
"eval_gate2_f1": 0.7921038946740504,
|
| 298 |
+
"eval_gate2_precision": 0.802926749773104,
|
| 299 |
+
"eval_gate2_recall": 0.7855851424596473,
|
| 300 |
"eval_gate2_support": 5480,
|
| 301 |
+
"eval_gate3_accuracy": 0.8915610599078341,
|
| 302 |
+
"eval_gate3_f1": 0.8022192601674913,
|
| 303 |
+
"eval_gate3_precision": 0.7867252072690095,
|
| 304 |
+
"eval_gate3_recall": 0.8274418766924675,
|
| 305 |
+
"eval_gate3_support": 6944,
|
| 306 |
+
"eval_loss": 1.0077552795410156,
|
| 307 |
+
"eval_runtime": 9.7597,
|
| 308 |
+
"eval_samples_per_second": 1272.994,
|
| 309 |
+
"eval_steps_per_second": 39.858,
|
| 310 |
+
"step": 31070
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 10.0,
|
| 314 |
+
"step": 31070,
|
| 315 |
+
"total_flos": 6.336727068480365e+16,
|
| 316 |
+
"train_loss": 0.3354094574986296,
|
| 317 |
+
"train_runtime": 2265.1995,
|
| 318 |
+
"train_samples_per_second": 438.787,
|
| 319 |
+
"train_steps_per_second": 13.716
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 10.0,
|
| 323 |
+
"eval_gate1_accuracy": 0.9231751824817518,
|
| 324 |
+
"eval_gate1_f1": 0.91874922234221,
|
| 325 |
+
"eval_gate1_precision": 0.9213536363863121,
|
| 326 |
+
"eval_gate1_recall": 0.916446997689221,
|
| 327 |
"eval_gate1_support": 5480,
|
| 328 |
+
"eval_gate2_accuracy": 0.8593065693430657,
|
| 329 |
+
"eval_gate2_f1": 0.7763067392280264,
|
| 330 |
+
"eval_gate2_precision": 0.7921477517391428,
|
| 331 |
+
"eval_gate2_recall": 0.775442797363134,
|
| 332 |
"eval_gate2_support": 5480,
|
| 333 |
+
"eval_gate3_accuracy": 0.8810483870967742,
|
| 334 |
+
"eval_gate3_f1": 0.778218086166726,
|
| 335 |
+
"eval_gate3_precision": 0.75635112362015,
|
| 336 |
+
"eval_gate3_recall": 0.8147429373450497,
|
| 337 |
+
"eval_gate3_support": 6944,
|
| 338 |
+
"eval_loss": 0.7152989506721497,
|
| 339 |
+
"eval_runtime": 9.7809,
|
| 340 |
+
"eval_samples_per_second": 1270.237,
|
| 341 |
+
"eval_steps_per_second": 39.772,
|
| 342 |
+
"step": 31070
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 10.0,
|
| 346 |
+
"eval_gate1_accuracy": 0.9204524721766101,
|
| 347 |
+
"eval_gate1_f1": 0.9175696677724912,
|
| 348 |
+
"eval_gate1_precision": 0.9198718411071631,
|
| 349 |
+
"eval_gate1_recall": 0.9156173695177233,
|
| 350 |
"eval_gate1_support": 5481,
|
| 351 |
+
"eval_gate2_accuracy": 0.8573253056011677,
|
| 352 |
+
"eval_gate2_f1": 0.7662196765345953,
|
| 353 |
+
"eval_gate2_precision": 0.7540255243234163,
|
| 354 |
+
"eval_gate2_recall": 0.7899899686908007,
|
| 355 |
"eval_gate2_support": 5481,
|
| 356 |
+
"eval_gate3_accuracy": 0.8761520737327189,
|
| 357 |
+
"eval_gate3_f1": 0.7204269422018509,
|
| 358 |
+
"eval_gate3_precision": 0.7065840948237819,
|
| 359 |
+
"eval_gate3_recall": 0.7411799394675562,
|
| 360 |
+
"eval_gate3_support": 6944,
|
| 361 |
+
"eval_loss": 0.7167699933052063,
|
| 362 |
+
"eval_runtime": 9.951,
|
| 363 |
+
"eval_samples_per_second": 1248.614,
|
| 364 |
+
"eval_steps_per_second": 39.091,
|
| 365 |
+
"step": 31070
|
| 366 |
}
|
| 367 |
],
|
| 368 |
"logging_steps": 500,
|
| 369 |
+
"max_steps": 31070,
|
| 370 |
"num_input_tokens_seen": 0,
|
| 371 |
"num_train_epochs": 10,
|
| 372 |
"save_steps": 500,
|
|
|
|
| 382 |
"attributes": {}
|
| 383 |
}
|
| 384 |
},
|
| 385 |
+
"total_flos": 6.336727068480365e+16,
|
| 386 |
"train_batch_size": 32,
|
| 387 |
"trial_name": null,
|
| 388 |
"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 5841
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f7fd5367a51397aa9e61e2f6c06b3f993347adf643ea400b270a4147e24f944
|
| 3 |
size 5841
|