Training in progress, epoch 0
Browse files- all_results.json +6 -6
- model.safetensors +1 -1
- runs/May17_01-25-40_MSI/events.out.tfevents.1715919261.MSI.2780.1 +2 -2
- runs/May17_11-05-38_MSI/events.out.tfevents.1715937182.MSI.31856.0 +3 -0
- runs/May20_21-23-43_MSI/events.out.tfevents.1716233354.MSI.16400.0 +3 -0
- test_results.json +6 -6
- trainer_state.json +434 -89
- training_args.bin +1 -1
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 0.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.32972972972973,
|
| 3 |
+
"eval_accuracy": 0.8530805687203792,
|
| 4 |
+
"eval_loss": 0.5416910648345947,
|
| 5 |
+
"eval_runtime": 510.4742,
|
| 6 |
+
"eval_samples_per_second": 0.413,
|
| 7 |
+
"eval_steps_per_second": 0.104
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 344943528
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1601a546b99af2829e85db83871cd2a514d5723481dd3c2e59bb50fcf61aeefa
|
| 3 |
size 344943528
|
runs/May17_01-25-40_MSI/events.out.tfevents.1715919261.MSI.2780.1
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:ed0982addda4b018181715c03bc06d1b440e40f688999fa08fa3fec3b3fd2190
|
| 3 |
+
size 1703
|
runs/May17_11-05-38_MSI/events.out.tfevents.1715937182.MSI.31856.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c508a648698188b60860af4350f0800ad3035409e88d6b27ce941ba4adaf3230
|
| 3 |
+
size 346
|
runs/May20_21-23-43_MSI/events.out.tfevents.1716233354.MSI.16400.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:71dd945d78a184eb77c489b6627992e1e10efd55d88c5f067e2ac0b68f7c6398
|
| 3 |
+
size 9400
|
test_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 0.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.32972972972973,
|
| 3 |
+
"eval_accuracy": 0.8530805687203792,
|
| 4 |
+
"eval_loss": 0.5416910648345947,
|
| 5 |
+
"eval_runtime": 510.4742,
|
| 6 |
+
"eval_samples_per_second": 0.413,
|
| 7 |
+
"eval_steps_per_second": 0.104
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,141 +1,486 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "videomae-base-finetuned-2\\checkpoint-
|
| 4 |
-
"epoch":
|
| 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,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss":
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss":
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss":
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"
|
| 35 |
-
"
|
| 36 |
-
"
|
| 37 |
-
"eval_samples_per_second": 0.365,
|
| 38 |
-
"eval_steps_per_second": 0.093,
|
| 39 |
-
"step": 38
|
| 40 |
-
},
|
| 41 |
-
{
|
| 42 |
-
"epoch": 1.027027027027027,
|
| 43 |
-
"grad_norm": 6.076575756072998,
|
| 44 |
-
"learning_rate": 2.575757575757576e-05,
|
| 45 |
-
"loss": 0.6072,
|
| 46 |
"step": 40
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"epoch":
|
| 50 |
-
"grad_norm":
|
| 51 |
-
"learning_rate":
|
| 52 |
-
"loss":
|
| 53 |
"step": 50
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch":
|
| 57 |
-
"grad_norm":
|
| 58 |
-
"learning_rate":
|
| 59 |
-
"loss":
|
| 60 |
"step": 60
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"epoch":
|
| 64 |
-
"grad_norm":
|
| 65 |
-
"learning_rate":
|
| 66 |
-
"loss":
|
| 67 |
"step": 70
|
| 68 |
},
|
| 69 |
{
|
| 70 |
-
"epoch":
|
| 71 |
-
"
|
| 72 |
-
"
|
| 73 |
-
"
|
| 74 |
-
"
|
| 75 |
-
|
| 76 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 77 |
},
|
| 78 |
{
|
| 79 |
-
"epoch":
|
| 80 |
-
"
|
| 81 |
-
"
|
| 82 |
-
"
|
| 83 |
-
"
|
| 84 |
-
"train_samples_per_second": 0.133,
|
| 85 |
-
"train_steps_per_second": 0.033
|
| 86 |
},
|
| 87 |
{
|
| 88 |
-
"epoch":
|
| 89 |
-
"
|
| 90 |
-
"
|
| 91 |
-
"
|
| 92 |
-
"
|
| 93 |
-
|
| 94 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 95 |
},
|
| 96 |
{
|
| 97 |
-
"epoch":
|
| 98 |
-
"eval_accuracy": 0.
|
| 99 |
-
"eval_loss": 0.
|
| 100 |
-
"eval_runtime":
|
| 101 |
-
"eval_samples_per_second": 0.
|
| 102 |
"eval_steps_per_second": 0.1,
|
| 103 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 104 |
},
|
| 105 |
{
|
| 106 |
-
"epoch":
|
| 107 |
-
"eval_accuracy": 0.
|
| 108 |
-
"eval_loss": 0.
|
| 109 |
-
"eval_runtime":
|
| 110 |
"eval_samples_per_second": 0.403,
|
| 111 |
"eval_steps_per_second": 0.103,
|
| 112 |
-
"step":
|
| 113 |
},
|
| 114 |
{
|
| 115 |
-
"epoch":
|
| 116 |
-
"eval_accuracy": 0.
|
| 117 |
-
"eval_loss": 0.
|
| 118 |
-
"eval_runtime":
|
| 119 |
-
"eval_samples_per_second": 0.
|
| 120 |
-
"eval_steps_per_second": 0.
|
| 121 |
-
"step":
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch":
|
| 125 |
-
"eval_accuracy": 0.
|
| 126 |
-
"eval_loss": 0.
|
| 127 |
-
"eval_runtime":
|
| 128 |
-
"eval_samples_per_second": 0.
|
| 129 |
-
"eval_steps_per_second": 0.
|
| 130 |
-
"step":
|
| 131 |
}
|
| 132 |
],
|
| 133 |
"logging_steps": 10,
|
| 134 |
-
"max_steps":
|
| 135 |
"num_input_tokens_seen": 0,
|
| 136 |
"num_train_epochs": 9223372036854775807,
|
| 137 |
"save_steps": 500,
|
| 138 |
-
"total_flos":
|
| 139 |
"train_batch_size": 4,
|
| 140 |
"trial_name": null,
|
| 141 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8636363636363636,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-2\\checkpoint-555",
|
| 4 |
+
"epoch": 2.32972972972973,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 555,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.018018018018018018,
|
| 13 |
+
"grad_norm": 9.139641761779785,
|
| 14 |
+
"learning_rate": 8.92857142857143e-06,
|
| 15 |
+
"loss": 1.3861,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.036036036036036036,
|
| 20 |
+
"grad_norm": 8.826565742492676,
|
| 21 |
+
"learning_rate": 1.785714285714286e-05,
|
| 22 |
+
"loss": 1.2321,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.05405405405405406,
|
| 27 |
+
"grad_norm": 6.707639217376709,
|
| 28 |
+
"learning_rate": 2.6785714285714288e-05,
|
| 29 |
+
"loss": 1.243,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.07207207207207207,
|
| 34 |
+
"grad_norm": 11.348919868469238,
|
| 35 |
+
"learning_rate": 3.571428571428572e-05,
|
| 36 |
+
"loss": 1.1312,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.09009009009009009,
|
| 41 |
+
"grad_norm": 14.494072914123535,
|
| 42 |
+
"learning_rate": 4.464285714285715e-05,
|
| 43 |
+
"loss": 1.3058,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.10810810810810811,
|
| 48 |
+
"grad_norm": 7.667250156402588,
|
| 49 |
+
"learning_rate": 4.9599198396793586e-05,
|
| 50 |
+
"loss": 1.2019,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.12612612612612611,
|
| 55 |
+
"grad_norm": 8.06010627746582,
|
| 56 |
+
"learning_rate": 4.859719438877756e-05,
|
| 57 |
+
"loss": 1.2813,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.14414414414414414,
|
| 62 |
+
"grad_norm": 10.494473457336426,
|
| 63 |
+
"learning_rate": 4.7595190380761524e-05,
|
| 64 |
+
"loss": 1.2292,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.16216216216216217,
|
| 69 |
+
"grad_norm": 11.735258102416992,
|
| 70 |
+
"learning_rate": 4.6593186372745496e-05,
|
| 71 |
+
"loss": 1.2424,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.18018018018018017,
|
| 76 |
+
"grad_norm": 7.723865509033203,
|
| 77 |
+
"learning_rate": 4.559118236472946e-05,
|
| 78 |
+
"loss": 1.2314,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.1981981981981982,
|
| 83 |
+
"grad_norm": 8.27652645111084,
|
| 84 |
+
"learning_rate": 4.458917835671343e-05,
|
| 85 |
+
"loss": 1.0673,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.21621621621621623,
|
| 90 |
+
"grad_norm": 7.197485446929932,
|
| 91 |
+
"learning_rate": 4.358717434869739e-05,
|
| 92 |
+
"loss": 1.337,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.23423423423423423,
|
| 97 |
+
"grad_norm": 9.621479988098145,
|
| 98 |
+
"learning_rate": 4.2585170340681366e-05,
|
| 99 |
+
"loss": 1.2103,
|
| 100 |
+
"step": 130
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.25225225225225223,
|
| 104 |
+
"grad_norm": 8.090988159179688,
|
| 105 |
+
"learning_rate": 4.158316633266534e-05,
|
| 106 |
+
"loss": 1.1458,
|
| 107 |
+
"step": 140
|
|
|
|
|
|
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.2702702702702703,
|
| 111 |
+
"grad_norm": 8.642086029052734,
|
| 112 |
+
"learning_rate": 4.0581162324649304e-05,
|
| 113 |
+
"loss": 1.0949,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2882882882882883,
|
| 118 |
+
"grad_norm": 6.245614051818848,
|
| 119 |
+
"learning_rate": 3.957915831663327e-05,
|
| 120 |
+
"loss": 0.9812,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3063063063063063,
|
| 125 |
+
"grad_norm": 10.789504051208496,
|
| 126 |
+
"learning_rate": 3.8577154308617235e-05,
|
| 127 |
+
"loss": 0.9814,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.32432432432432434,
|
| 132 |
+
"grad_norm": 17.914480209350586,
|
| 133 |
+
"learning_rate": 3.75751503006012e-05,
|
| 134 |
+
"loss": 0.8667,
|
| 135 |
+
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.33513513513513515,
|
| 139 |
+
"eval_accuracy": 0.5727272727272728,
|
| 140 |
+
"eval_loss": 0.9740895628929138,
|
| 141 |
+
"eval_runtime": 280.3552,
|
| 142 |
+
"eval_samples_per_second": 0.392,
|
| 143 |
"eval_steps_per_second": 0.1,
|
| 144 |
+
"step": 186
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.0072072072072071,
|
| 148 |
+
"grad_norm": 12.168312072753906,
|
| 149 |
+
"learning_rate": 3.657314629258517e-05,
|
| 150 |
+
"loss": 0.9448,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.0252252252252252,
|
| 155 |
+
"grad_norm": 7.322641849517822,
|
| 156 |
+
"learning_rate": 3.5571142284569145e-05,
|
| 157 |
+
"loss": 1.1193,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.0432432432432432,
|
| 162 |
+
"grad_norm": 12.514275550842285,
|
| 163 |
+
"learning_rate": 3.456913827655311e-05,
|
| 164 |
+
"loss": 0.8378,
|
| 165 |
+
"step": 210
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.0612612612612613,
|
| 169 |
+
"grad_norm": 7.457886695861816,
|
| 170 |
+
"learning_rate": 3.356713426853708e-05,
|
| 171 |
+
"loss": 0.7454,
|
| 172 |
+
"step": 220
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.0792792792792794,
|
| 176 |
+
"grad_norm": 11.546749114990234,
|
| 177 |
+
"learning_rate": 3.256513026052104e-05,
|
| 178 |
+
"loss": 0.8304,
|
| 179 |
+
"step": 230
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.0972972972972972,
|
| 183 |
+
"grad_norm": 10.54076099395752,
|
| 184 |
+
"learning_rate": 3.156312625250501e-05,
|
| 185 |
+
"loss": 1.0602,
|
| 186 |
+
"step": 240
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.1153153153153152,
|
| 190 |
+
"grad_norm": 31.803747177124023,
|
| 191 |
+
"learning_rate": 3.0561122244488974e-05,
|
| 192 |
+
"loss": 0.806,
|
| 193 |
+
"step": 250
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 1.1333333333333333,
|
| 197 |
+
"grad_norm": 28.31694793701172,
|
| 198 |
+
"learning_rate": 2.955911823647295e-05,
|
| 199 |
+
"loss": 0.8338,
|
| 200 |
+
"step": 260
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.1513513513513514,
|
| 204 |
+
"grad_norm": 9.744572639465332,
|
| 205 |
+
"learning_rate": 2.8557114228456915e-05,
|
| 206 |
+
"loss": 0.8696,
|
| 207 |
+
"step": 270
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.1693693693693694,
|
| 211 |
+
"grad_norm": 12.256711959838867,
|
| 212 |
+
"learning_rate": 2.7555110220440884e-05,
|
| 213 |
+
"loss": 0.7245,
|
| 214 |
+
"step": 280
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 1.1873873873873875,
|
| 218 |
+
"grad_norm": 7.700608730316162,
|
| 219 |
+
"learning_rate": 2.655310621242485e-05,
|
| 220 |
+
"loss": 0.6561,
|
| 221 |
+
"step": 290
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 1.2054054054054055,
|
| 225 |
+
"grad_norm": 14.277639389038086,
|
| 226 |
+
"learning_rate": 2.5551102204408815e-05,
|
| 227 |
+
"loss": 0.7837,
|
| 228 |
+
"step": 300
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.2234234234234234,
|
| 232 |
+
"grad_norm": 9.197588920593262,
|
| 233 |
+
"learning_rate": 2.4549098196392788e-05,
|
| 234 |
+
"loss": 0.5502,
|
| 235 |
+
"step": 310
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 1.2414414414414414,
|
| 239 |
+
"grad_norm": 3.099351167678833,
|
| 240 |
+
"learning_rate": 2.3547094188376757e-05,
|
| 241 |
+
"loss": 0.5195,
|
| 242 |
+
"step": 320
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 1.2594594594594595,
|
| 246 |
+
"grad_norm": 9.323990821838379,
|
| 247 |
+
"learning_rate": 2.2545090180360722e-05,
|
| 248 |
+
"loss": 1.1324,
|
| 249 |
+
"step": 330
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.2774774774774775,
|
| 253 |
+
"grad_norm": 10.672691345214844,
|
| 254 |
+
"learning_rate": 2.1543086172344688e-05,
|
| 255 |
+
"loss": 0.4475,
|
| 256 |
+
"step": 340
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.2954954954954956,
|
| 260 |
+
"grad_norm": 0.755842924118042,
|
| 261 |
+
"learning_rate": 2.054108216432866e-05,
|
| 262 |
+
"loss": 0.5673,
|
| 263 |
+
"step": 350
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.3135135135135134,
|
| 267 |
+
"grad_norm": 1.0543122291564941,
|
| 268 |
+
"learning_rate": 1.9539078156312626e-05,
|
| 269 |
+
"loss": 0.7184,
|
| 270 |
+
"step": 360
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.3315315315315315,
|
| 274 |
+
"grad_norm": 12.980474472045898,
|
| 275 |
+
"learning_rate": 1.8537074148296592e-05,
|
| 276 |
+
"loss": 0.8111,
|
| 277 |
+
"step": 370
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 1.3351351351351353,
|
| 281 |
+
"eval_accuracy": 0.7363636363636363,
|
| 282 |
+
"eval_loss": 0.8580251336097717,
|
| 283 |
+
"eval_runtime": 277.5592,
|
| 284 |
+
"eval_samples_per_second": 0.396,
|
| 285 |
+
"eval_steps_per_second": 0.101,
|
| 286 |
+
"step": 372
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 2.0144144144144143,
|
| 290 |
+
"grad_norm": 15.273543357849121,
|
| 291 |
+
"learning_rate": 1.7535070140280564e-05,
|
| 292 |
+
"loss": 0.3997,
|
| 293 |
+
"step": 380
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 2.0324324324324325,
|
| 297 |
+
"grad_norm": 9.675086975097656,
|
| 298 |
+
"learning_rate": 1.653306613226453e-05,
|
| 299 |
+
"loss": 0.2966,
|
| 300 |
+
"step": 390
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 2.0504504504504504,
|
| 304 |
+
"grad_norm": 0.6464720368385315,
|
| 305 |
+
"learning_rate": 1.5531062124248495e-05,
|
| 306 |
+
"loss": 0.3275,
|
| 307 |
+
"step": 400
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 2.0684684684684687,
|
| 311 |
+
"grad_norm": 5.569362163543701,
|
| 312 |
+
"learning_rate": 1.4529058116232466e-05,
|
| 313 |
+
"loss": 0.1516,
|
| 314 |
+
"step": 410
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.0864864864864865,
|
| 318 |
+
"grad_norm": 20.961273193359375,
|
| 319 |
+
"learning_rate": 1.3527054108216433e-05,
|
| 320 |
+
"loss": 0.5888,
|
| 321 |
+
"step": 420
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.1045045045045043,
|
| 325 |
+
"grad_norm": 27.309669494628906,
|
| 326 |
+
"learning_rate": 1.25250501002004e-05,
|
| 327 |
+
"loss": 0.4586,
|
| 328 |
+
"step": 430
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 2.1225225225225226,
|
| 332 |
+
"grad_norm": 2.0433521270751953,
|
| 333 |
+
"learning_rate": 1.152304609218437e-05,
|
| 334 |
+
"loss": 0.3667,
|
| 335 |
+
"step": 440
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 2.1405405405405404,
|
| 339 |
+
"grad_norm": 0.46590688824653625,
|
| 340 |
+
"learning_rate": 1.0521042084168337e-05,
|
| 341 |
+
"loss": 0.2505,
|
| 342 |
+
"step": 450
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 2.1585585585585587,
|
| 346 |
+
"grad_norm": 0.8750420808792114,
|
| 347 |
+
"learning_rate": 9.519038076152304e-06,
|
| 348 |
+
"loss": 0.2555,
|
| 349 |
+
"step": 460
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 2.1765765765765765,
|
| 353 |
+
"grad_norm": 0.5527063608169556,
|
| 354 |
+
"learning_rate": 8.517034068136273e-06,
|
| 355 |
+
"loss": 0.2616,
|
| 356 |
+
"step": 470
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 2.1945945945945944,
|
| 360 |
+
"grad_norm": 2.8728528022766113,
|
| 361 |
+
"learning_rate": 7.515030060120242e-06,
|
| 362 |
+
"loss": 0.1438,
|
| 363 |
+
"step": 480
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.2126126126126127,
|
| 367 |
+
"grad_norm": 1.5662002563476562,
|
| 368 |
+
"learning_rate": 6.513026052104208e-06,
|
| 369 |
+
"loss": 0.5138,
|
| 370 |
+
"step": 490
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 2.2306306306306305,
|
| 374 |
+
"grad_norm": 70.13788604736328,
|
| 375 |
+
"learning_rate": 5.511022044088176e-06,
|
| 376 |
+
"loss": 0.5573,
|
| 377 |
+
"step": 500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 2.2486486486486488,
|
| 381 |
+
"grad_norm": 2.147181510925293,
|
| 382 |
+
"learning_rate": 4.5090180360721445e-06,
|
| 383 |
+
"loss": 0.2309,
|
| 384 |
+
"step": 510
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.2666666666666666,
|
| 388 |
+
"grad_norm": 73.13829040527344,
|
| 389 |
+
"learning_rate": 3.5070140280561122e-06,
|
| 390 |
+
"loss": 0.2092,
|
| 391 |
+
"step": 520
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 2.2846846846846844,
|
| 395 |
+
"grad_norm": 33.50334167480469,
|
| 396 |
+
"learning_rate": 2.5050100200400804e-06,
|
| 397 |
+
"loss": 0.2009,
|
| 398 |
+
"step": 530
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 2.3027027027027027,
|
| 402 |
+
"grad_norm": 58.736778259277344,
|
| 403 |
+
"learning_rate": 1.5030060120240482e-06,
|
| 404 |
+
"loss": 0.2882,
|
| 405 |
+
"step": 540
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 2.3207207207207206,
|
| 409 |
+
"grad_norm": 7.180970191955566,
|
| 410 |
+
"learning_rate": 5.01002004008016e-07,
|
| 411 |
+
"loss": 0.4008,
|
| 412 |
+
"step": 550
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 2.32972972972973,
|
| 416 |
+
"eval_accuracy": 0.8636363636363636,
|
| 417 |
+
"eval_loss": 0.5939798355102539,
|
| 418 |
+
"eval_runtime": 273.4292,
|
| 419 |
+
"eval_samples_per_second": 0.402,
|
| 420 |
+
"eval_steps_per_second": 0.102,
|
| 421 |
+
"step": 555
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 2.32972972972973,
|
| 425 |
+
"step": 555,
|
| 426 |
+
"total_flos": 2.758836207590572e+18,
|
| 427 |
+
"train_loss": 0.7599288858808912,
|
| 428 |
+
"train_runtime": 12937.383,
|
| 429 |
+
"train_samples_per_second": 0.172,
|
| 430 |
+
"train_steps_per_second": 0.043
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 2.32972972972973,
|
| 434 |
+
"eval_accuracy": 0.9487179487179487,
|
| 435 |
+
"eval_loss": 0.15456415712833405,
|
| 436 |
+
"eval_runtime": 1766.8377,
|
| 437 |
+
"eval_samples_per_second": 0.419,
|
| 438 |
+
"eval_steps_per_second": 0.105,
|
| 439 |
+
"step": 555
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 2.32972972972973,
|
| 443 |
+
"eval_accuracy": 0.8530805687203792,
|
| 444 |
+
"eval_loss": 0.54169100522995,
|
| 445 |
+
"eval_runtime": 505.2243,
|
| 446 |
+
"eval_samples_per_second": 0.418,
|
| 447 |
+
"eval_steps_per_second": 0.105,
|
| 448 |
+
"step": 555
|
| 449 |
},
|
| 450 |
{
|
| 451 |
+
"epoch": 2.32972972972973,
|
| 452 |
+
"eval_accuracy": 0.8636363636363636,
|
| 453 |
+
"eval_loss": 0.5939798355102539,
|
| 454 |
+
"eval_runtime": 272.8909,
|
| 455 |
"eval_samples_per_second": 0.403,
|
| 456 |
"eval_steps_per_second": 0.103,
|
| 457 |
+
"step": 555
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 2.32972972972973,
|
| 461 |
+
"eval_accuracy": 0.8636363636363636,
|
| 462 |
+
"eval_loss": 0.5939797163009644,
|
| 463 |
+
"eval_runtime": 262.8979,
|
| 464 |
+
"eval_samples_per_second": 0.418,
|
| 465 |
+
"eval_steps_per_second": 0.107,
|
| 466 |
+
"step": 555
|
| 467 |
},
|
| 468 |
{
|
| 469 |
+
"epoch": 2.32972972972973,
|
| 470 |
+
"eval_accuracy": 0.8530805687203792,
|
| 471 |
+
"eval_loss": 0.5416910648345947,
|
| 472 |
+
"eval_runtime": 510.4742,
|
| 473 |
+
"eval_samples_per_second": 0.413,
|
| 474 |
+
"eval_steps_per_second": 0.104,
|
| 475 |
+
"step": 555
|
| 476 |
}
|
| 477 |
],
|
| 478 |
"logging_steps": 10,
|
| 479 |
+
"max_steps": 555,
|
| 480 |
"num_input_tokens_seen": 0,
|
| 481 |
"num_train_epochs": 9223372036854775807,
|
| 482 |
"save_steps": 500,
|
| 483 |
+
"total_flos": 2.758836207590572e+18,
|
| 484 |
"train_batch_size": 4,
|
| 485 |
"trial_name": null,
|
| 486 |
"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:a9b96b3f06bc1b4bc605b44ff522e3676ebc6b85fccbe4e3870880b7b91d1ddc
|
| 3 |
size 4984
|