Upload logs/train_log_Llama-3.1-8B_seed999.json with huggingface_hub
Browse files
logs/train_log_Llama-3.1-8B_seed999.json
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"loss": 2.3616,
|
| 4 |
+
"grad_norm": 1.9140347242355347,
|
| 5 |
+
"learning_rate": 0.00018,
|
| 6 |
+
"epoch": 0.5,
|
| 7 |
+
"step": 10
|
| 8 |
+
},
|
| 9 |
+
{
|
| 10 |
+
"loss": 0.4068,
|
| 11 |
+
"grad_norm": 1.4040199518203735,
|
| 12 |
+
"learning_rate": 0.00018,
|
| 13 |
+
"epoch": 1.0,
|
| 14 |
+
"step": 20
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"loss": 0.183,
|
| 18 |
+
"grad_norm": 0.5609062910079956,
|
| 19 |
+
"learning_rate": 0.0001577777777777778,
|
| 20 |
+
"epoch": 1.5,
|
| 21 |
+
"step": 30
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"loss": 0.1524,
|
| 25 |
+
"grad_norm": 0.8594111800193787,
|
| 26 |
+
"learning_rate": 0.00013555555555555556,
|
| 27 |
+
"epoch": 2.0,
|
| 28 |
+
"step": 40
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"loss": 0.1158,
|
| 32 |
+
"grad_norm": 0.6145699620246887,
|
| 33 |
+
"learning_rate": 0.00011333333333333334,
|
| 34 |
+
"epoch": 2.5,
|
| 35 |
+
"step": 50
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"loss": 0.1002,
|
| 39 |
+
"grad_norm": 0.6389060020446777,
|
| 40 |
+
"learning_rate": 9.111111111111112e-05,
|
| 41 |
+
"epoch": 3.0,
|
| 42 |
+
"step": 60
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"loss": 0.0852,
|
| 46 |
+
"grad_norm": 0.5036048889160156,
|
| 47 |
+
"learning_rate": 6.88888888888889e-05,
|
| 48 |
+
"epoch": 3.5,
|
| 49 |
+
"step": 70
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"loss": 0.0608,
|
| 53 |
+
"grad_norm": 0.6369991302490234,
|
| 54 |
+
"learning_rate": 4.666666666666667e-05,
|
| 55 |
+
"epoch": 4.0,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"loss": 0.0483,
|
| 60 |
+
"grad_norm": 0.6636818051338196,
|
| 61 |
+
"learning_rate": 2.4444444444444445e-05,
|
| 62 |
+
"epoch": 4.5,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"loss": 0.0469,
|
| 67 |
+
"grad_norm": 0.8848990201950073,
|
| 68 |
+
"learning_rate": 2.2222222222222225e-06,
|
| 69 |
+
"epoch": 5.0,
|
| 70 |
+
"step": 100
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"train_runtime": 349.3684,
|
| 74 |
+
"train_samples_per_second": 17.975,
|
| 75 |
+
"train_steps_per_second": 0.286,
|
| 76 |
+
"total_flos": 5.098583944986624e+16,
|
| 77 |
+
"train_loss": 0.35608491629362105,
|
| 78 |
+
"epoch": 5.0,
|
| 79 |
+
"step": 100
|
| 80 |
+
}
|
| 81 |
+
]
|