latest = step 6159 (nDCG@10=0.4885)
Browse files- trainer_state.json +46 -6
trainer_state.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 0.
|
| 4 |
-
"best_model_checkpoint": "checkpoints/med-ruri-v3-70m-v2-from-ruri/checkpoint-
|
| 5 |
-
"epoch":
|
| 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,
|
|
@@ -1076,6 +1076,46 @@
|
|
| 1076 |
"eval_samples_per_second": 0.0,
|
| 1077 |
"eval_steps_per_second": 0.0,
|
| 1078 |
"step": 6000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1079 |
}
|
| 1080 |
],
|
| 1081 |
"logging_steps": 50,
|
|
@@ -1090,7 +1130,7 @@
|
|
| 1090 |
"should_evaluate": false,
|
| 1091 |
"should_log": false,
|
| 1092 |
"should_save": true,
|
| 1093 |
-
"should_training_stop":
|
| 1094 |
},
|
| 1095 |
"attributes": {}
|
| 1096 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 6159,
|
| 3 |
+
"best_metric": 0.4884732005383661,
|
| 4 |
+
"best_model_checkpoint": "checkpoints/med-ruri-v3-70m-v2-from-ruri/checkpoint-6159",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 6159,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 1076 |
"eval_samples_per_second": 0.0,
|
| 1077 |
"eval_steps_per_second": 0.0,
|
| 1078 |
"step": 6000
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 0.9823023218054879,
|
| 1082 |
+
"grad_norm": 26.53546714782715,
|
| 1083 |
+
"learning_rate": 1.8800205093146473e-07,
|
| 1084 |
+
"loss": 0.4033943176269531,
|
| 1085 |
+
"step": 6050
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 0.9904205228121449,
|
| 1089 |
+
"grad_norm": 32.0621452331543,
|
| 1090 |
+
"learning_rate": 1.025465732353444e-07,
|
| 1091 |
+
"loss": 0.4093938446044922,
|
| 1092 |
+
"step": 6100
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 0.9985387238188017,
|
| 1096 |
+
"grad_norm": 28.408830642700195,
|
| 1097 |
+
"learning_rate": 1.7091095539224067e-08,
|
| 1098 |
+
"loss": 0.39267391204833985,
|
| 1099 |
+
"step": 6150
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 1.0,
|
| 1103 |
+
"eval_med-ir_cosine_accuracy@1": 0.27,
|
| 1104 |
+
"eval_med-ir_cosine_accuracy@10": 0.735,
|
| 1105 |
+
"eval_med-ir_cosine_accuracy@5": 0.609,
|
| 1106 |
+
"eval_med-ir_cosine_map@10": 0.41104007936507936,
|
| 1107 |
+
"eval_med-ir_cosine_mrr@10": 0.4110400793650802,
|
| 1108 |
+
"eval_med-ir_cosine_ndcg@10": 0.4884732005383661,
|
| 1109 |
+
"eval_med-ir_cosine_precision@1": 0.27,
|
| 1110 |
+
"eval_med-ir_cosine_precision@10": 0.0735,
|
| 1111 |
+
"eval_med-ir_cosine_precision@5": 0.12179999999999998,
|
| 1112 |
+
"eval_med-ir_cosine_recall@1": 0.27,
|
| 1113 |
+
"eval_med-ir_cosine_recall@10": 0.735,
|
| 1114 |
+
"eval_med-ir_cosine_recall@5": 0.609,
|
| 1115 |
+
"eval_runtime": 53.022,
|
| 1116 |
+
"eval_samples_per_second": 0.0,
|
| 1117 |
+
"eval_steps_per_second": 0.0,
|
| 1118 |
+
"step": 6159
|
| 1119 |
}
|
| 1120 |
],
|
| 1121 |
"logging_steps": 50,
|
|
|
|
| 1130 |
"should_evaluate": false,
|
| 1131 |
"should_log": false,
|
| 1132 |
"should_save": true,
|
| 1133 |
+
"should_training_stop": true
|
| 1134 |
},
|
| 1135 |
"attributes": {}
|
| 1136 |
}
|