Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- checkpoints-d2.0-t1/checkpoint-17408/eval_state.json +3 -0
- checkpoints-d2.0-t1/checkpoint-17408/model.safetensors +3 -0
- checkpoints-d2.0-t1/checkpoint-17408/optimizer.pt +3 -0
- checkpoints-d2.0-t1/checkpoint-17408/rng_state.pth +3 -0
- checkpoints-d2.0-t1/checkpoint-17408/scaler.pt +3 -0
- checkpoints-d2.0-t1/checkpoint-17408/scheduler.pt +3 -0
- checkpoints-d2.0-t1/checkpoint-17408/trainer_state.json +833 -0
- checkpoints-d2.0-t1/checkpoint-17408/training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -79,3 +79,4 @@ checkpoints-d1.2/checkpoint-20480/eval_state.json filter=lfs diff=lfs merge=lfs
|
|
| 79 |
checkpoints-v1.1-d/checkpoint-21504/eval_state.json filter=lfs diff=lfs merge=lfs -text
|
| 80 |
checkpoints-d2.0/checkpoint-7168/eval_state.json filter=lfs diff=lfs merge=lfs -text
|
| 81 |
checkpoints-d2.0/checkpoint-21504/eval_state.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 79 |
checkpoints-v1.1-d/checkpoint-21504/eval_state.json filter=lfs diff=lfs merge=lfs -text
|
| 80 |
checkpoints-d2.0/checkpoint-7168/eval_state.json filter=lfs diff=lfs merge=lfs -text
|
| 81 |
checkpoints-d2.0/checkpoint-21504/eval_state.json filter=lfs diff=lfs merge=lfs -text
|
| 82 |
+
checkpoints-d2.0-t1/checkpoint-17408/eval_state.json filter=lfs diff=lfs merge=lfs -text
|
checkpoints-d2.0-t1/checkpoint-17408/eval_state.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9444efb7d76e323b7b3178005a04d895fb24e07fab9707d99122f0f140b11f56
|
| 3 |
+
size 42402279
|
checkpoints-d2.0-t1/checkpoint-17408/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3da65479c29f71f785b44f33c17a8204bee782d86a25c2e8af7532867fccc21d
|
| 3 |
+
size 32318104
|
checkpoints-d2.0-t1/checkpoint-17408/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ad8e37cf61b6f2fd7cc202071742e70be39c400ecaa1929a97c9bde1a46273b
|
| 3 |
+
size 10010635
|
checkpoints-d2.0-t1/checkpoint-17408/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cf1f3265930d1f1ea49e43ac26f5b9dcff623d55dd3c8497a3326caee03b2129
|
| 3 |
+
size 14645
|
checkpoints-d2.0-t1/checkpoint-17408/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cde0a87150daa6e4c1711b6971f37b1f87e96bb470ca1bc2e66650b7bb9252d6
|
| 3 |
+
size 1383
|
checkpoints-d2.0-t1/checkpoint-17408/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:907bd363cbb9a2ad34bb5cd3891b5c4fef8b45659aaf375f17f2876c5871354f
|
| 3 |
+
size 1465
|
checkpoints-d2.0-t1/checkpoint-17408/trainer_state.json
ADDED
|
@@ -0,0 +1,833 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.8040275275968778,
|
| 6 |
+
"eval_steps": 1024,
|
| 7 |
+
"global_step": 17408,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.011823934229365849,
|
| 14 |
+
"grad_norm": 0.7295829057693481,
|
| 15 |
+
"learning_rate": 0.000498046875,
|
| 16 |
+
"loss": 1.8308284282684326,
|
| 17 |
+
"step": 256
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.023647868458731697,
|
| 21 |
+
"grad_norm": 0.8052719831466675,
|
| 22 |
+
"learning_rate": 0.000998046875,
|
| 23 |
+
"loss": 1.8338860273361206,
|
| 24 |
+
"step": 512
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03547180268809755,
|
| 28 |
+
"grad_norm": 0.8997901678085327,
|
| 29 |
+
"learning_rate": 0.000999640996023194,
|
| 30 |
+
"loss": 1.8343064785003662,
|
| 31 |
+
"step": 768
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.047295736917463395,
|
| 35 |
+
"grad_norm": 0.7250568270683289,
|
| 36 |
+
"learning_rate": 0.0009985588674043958,
|
| 37 |
+
"loss": 1.8296632766723633,
|
| 38 |
+
"step": 1024
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.047295736917463395,
|
| 42 |
+
"eval_beta_ce_loss": 0.6351920716022248,
|
| 43 |
+
"eval_bleu": 0.4998693474981933,
|
| 44 |
+
"eval_loss": 1.8211502122552428,
|
| 45 |
+
"eval_uni_ce_loss": 0.5507660694590443,
|
| 46 |
+
"step": 1024
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.047295736917463395,
|
| 50 |
+
"eval_beta_ce_loss": 0.6351920716022248,
|
| 51 |
+
"eval_bleu": 0.4998693474981933,
|
| 52 |
+
"eval_loss": 1.8211502122552428,
|
| 53 |
+
"eval_runtime": 139.8967,
|
| 54 |
+
"eval_samples_per_second": 200.098,
|
| 55 |
+
"eval_steps_per_second": 3.131,
|
| 56 |
+
"eval_uni_ce_loss": 0.5507660694590443,
|
| 57 |
+
"step": 1024
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.05911967114682925,
|
| 61 |
+
"grad_norm": 0.7675038576126099,
|
| 62 |
+
"learning_rate": 0.0009967551747861387,
|
| 63 |
+
"loss": 1.8191157579421997,
|
| 64 |
+
"step": 1280
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.0709436053761951,
|
| 68 |
+
"grad_norm": 0.8503870368003845,
|
| 69 |
+
"learning_rate": 0.000994232528651847,
|
| 70 |
+
"loss": 1.8177354335784912,
|
| 71 |
+
"step": 1536
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.08276753960556095,
|
| 75 |
+
"grad_norm": 0.8022050261497498,
|
| 76 |
+
"learning_rate": 0.0009909945800260092,
|
| 77 |
+
"loss": 1.807867407798767,
|
| 78 |
+
"step": 1792
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.09459147383492679,
|
| 82 |
+
"grad_norm": 0.9533084034919739,
|
| 83 |
+
"learning_rate": 0.0009870460151900522,
|
| 84 |
+
"loss": 1.8013904094696045,
|
| 85 |
+
"step": 2048
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.09459147383492679,
|
| 89 |
+
"eval_beta_ce_loss": 0.6264337543483194,
|
| 90 |
+
"eval_bleu": 0.49771606091961523,
|
| 91 |
+
"eval_loss": 1.7982131136606818,
|
| 92 |
+
"eval_uni_ce_loss": 0.545345604555792,
|
| 93 |
+
"step": 2048
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.09459147383492679,
|
| 97 |
+
"eval_beta_ce_loss": 0.6264337543483194,
|
| 98 |
+
"eval_bleu": 0.49771606091961523,
|
| 99 |
+
"eval_loss": 1.7982131136606818,
|
| 100 |
+
"eval_runtime": 133.1265,
|
| 101 |
+
"eval_samples_per_second": 210.274,
|
| 102 |
+
"eval_steps_per_second": 3.29,
|
| 103 |
+
"eval_uni_ce_loss": 0.545345604555792,
|
| 104 |
+
"step": 2048
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.10641540806429264,
|
| 108 |
+
"grad_norm": 0.9694677591323853,
|
| 109 |
+
"learning_rate": 0.0009823925488998885,
|
| 110 |
+
"loss": 1.8004993200302124,
|
| 111 |
+
"step": 2304
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.1182393422936585,
|
| 115 |
+
"grad_norm": 0.8076265454292297,
|
| 116 |
+
"learning_rate": 0.0009770409161149525,
|
| 117 |
+
"loss": 1.7975847721099854,
|
| 118 |
+
"step": 2560
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.13006327652302435,
|
| 122 |
+
"grad_norm": 0.7418428063392639,
|
| 123 |
+
"learning_rate": 0.0009709988622506973,
|
| 124 |
+
"loss": 1.789136528968811,
|
| 125 |
+
"step": 2816
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.1418872107523902,
|
| 129 |
+
"grad_norm": 0.7103869915008545,
|
| 130 |
+
"learning_rate": 0.000964275131968659,
|
| 131 |
+
"loss": 1.7754944562911987,
|
| 132 |
+
"step": 3072
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.1418872107523902,
|
| 136 |
+
"eval_beta_ce_loss": 0.620368564918161,
|
| 137 |
+
"eval_bleu": 0.49887627167388116,
|
| 138 |
+
"eval_loss": 1.7806649703413384,
|
| 139 |
+
"eval_uni_ce_loss": 0.5399278415936858,
|
| 140 |
+
"step": 3072
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.1418872107523902,
|
| 144 |
+
"eval_beta_ce_loss": 0.620368564918161,
|
| 145 |
+
"eval_bleu": 0.49887627167388116,
|
| 146 |
+
"eval_loss": 1.7806649703413384,
|
| 147 |
+
"eval_runtime": 134.7634,
|
| 148 |
+
"eval_samples_per_second": 207.72,
|
| 149 |
+
"eval_steps_per_second": 3.25,
|
| 150 |
+
"eval_uni_ce_loss": 0.5399278415936858,
|
| 151 |
+
"step": 3072
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.15371114498175603,
|
| 155 |
+
"grad_norm": 0.7851372957229614,
|
| 156 |
+
"learning_rate": 0.0009568794565203123,
|
| 157 |
+
"loss": 1.7792595624923706,
|
| 158 |
+
"step": 3328
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.1655350792111219,
|
| 162 |
+
"grad_norm": 0.7782461047172546,
|
| 163 |
+
"learning_rate": 0.0009488225396630347,
|
| 164 |
+
"loss": 1.7753657102584839,
|
| 165 |
+
"step": 3584
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.17735901344048774,
|
| 169 |
+
"grad_norm": 0.7312053442001343,
|
| 170 |
+
"learning_rate": 0.0009401160421685646,
|
| 171 |
+
"loss": 1.769174575805664,
|
| 172 |
+
"step": 3840
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.18918294766985358,
|
| 176 |
+
"grad_norm": 0.8110184669494629,
|
| 177 |
+
"learning_rate": 0.0009307725649463714,
|
| 178 |
+
"loss": 1.769570231437683,
|
| 179 |
+
"step": 4096
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.18918294766985358,
|
| 183 |
+
"eval_beta_ce_loss": 0.6151743780260217,
|
| 184 |
+
"eval_bleu": 0.5008352932587944,
|
| 185 |
+
"eval_loss": 1.766069833818636,
|
| 186 |
+
"eval_uni_ce_loss": 0.5357210769500906,
|
| 187 |
+
"step": 4096
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.18918294766985358,
|
| 191 |
+
"eval_beta_ce_loss": 0.6151743780260217,
|
| 192 |
+
"eval_bleu": 0.5008352932587944,
|
| 193 |
+
"eval_loss": 1.766069833818636,
|
| 194 |
+
"eval_runtime": 135.334,
|
| 195 |
+
"eval_samples_per_second": 206.844,
|
| 196 |
+
"eval_steps_per_second": 3.236,
|
| 197 |
+
"eval_uni_ce_loss": 0.5357210769500906,
|
| 198 |
+
"step": 4096
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.20100688189921945,
|
| 202 |
+
"grad_norm": 0.9300932288169861,
|
| 203 |
+
"learning_rate": 0.0009208056308063659,
|
| 204 |
+
"loss": 1.7624573707580566,
|
| 205 |
+
"step": 4352
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.2128308161285853,
|
| 209 |
+
"grad_norm": 0.7691043019294739,
|
| 210 |
+
"learning_rate": 0.0009102296648873445,
|
| 211 |
+
"loss": 1.7650220394134521,
|
| 212 |
+
"step": 4608
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.22465475035795113,
|
| 216 |
+
"grad_norm": 0.8132173418998718,
|
| 217 |
+
"learning_rate": 0.0008990599737794927,
|
| 218 |
+
"loss": 1.761474370956421,
|
| 219 |
+
"step": 4864
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.236478684587317,
|
| 223 |
+
"grad_norm": 0.7635871171951294,
|
| 224 |
+
"learning_rate": 0.0008873127233711644,
|
| 225 |
+
"loss": 1.7527934312820435,
|
| 226 |
+
"step": 5120
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.236478684587317,
|
| 230 |
+
"eval_beta_ce_loss": 0.611639610028158,
|
| 231 |
+
"eval_bleu": 0.5024511500541371,
|
| 232 |
+
"eval_loss": 1.753186773763944,
|
| 233 |
+
"eval_uni_ce_loss": 0.5299075521426658,
|
| 234 |
+
"step": 5120
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.236478684587317,
|
| 238 |
+
"eval_beta_ce_loss": 0.611639610028158,
|
| 239 |
+
"eval_bleu": 0.5024511500541371,
|
| 240 |
+
"eval_loss": 1.753186773763944,
|
| 241 |
+
"eval_runtime": 135.0739,
|
| 242 |
+
"eval_samples_per_second": 207.242,
|
| 243 |
+
"eval_steps_per_second": 3.243,
|
| 244 |
+
"eval_uni_ce_loss": 0.5299075521426658,
|
| 245 |
+
"step": 5120
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.24830261881668284,
|
| 249 |
+
"grad_norm": 0.7438808083534241,
|
| 250 |
+
"learning_rate": 0.0008750049154520011,
|
| 251 |
+
"loss": 1.7492367029190063,
|
| 252 |
+
"step": 5376
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.2601265530460487,
|
| 256 |
+
"grad_norm": 0.7799262404441833,
|
| 257 |
+
"learning_rate": 0.0008621543631062487,
|
| 258 |
+
"loss": 1.744637131690979,
|
| 259 |
+
"step": 5632
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.27195048727541454,
|
| 263 |
+
"grad_norm": 0.7694032192230225,
|
| 264 |
+
"learning_rate": 0.0008487796649318904,
|
| 265 |
+
"loss": 1.7494758367538452,
|
| 266 |
+
"step": 5888
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.2837744215047804,
|
| 270 |
+
"grad_norm": 0.7369454503059387,
|
| 271 |
+
"learning_rate": 0.0008349001781229053,
|
| 272 |
+
"loss": 1.7494639158248901,
|
| 273 |
+
"step": 6144
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.2837744215047804,
|
| 277 |
+
"eval_beta_ce_loss": 0.6058613158524309,
|
| 278 |
+
"eval_bleu": 0.5028032064859724,
|
| 279 |
+
"eval_loss": 1.739504432841523,
|
| 280 |
+
"eval_uni_ce_loss": 0.5277818014088287,
|
| 281 |
+
"step": 6144
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.2837744215047804,
|
| 285 |
+
"eval_beta_ce_loss": 0.6058613158524309,
|
| 286 |
+
"eval_bleu": 0.5028032064859724,
|
| 287 |
+
"eval_loss": 1.739504432841523,
|
| 288 |
+
"eval_runtime": 136.5059,
|
| 289 |
+
"eval_samples_per_second": 205.068,
|
| 290 |
+
"eval_steps_per_second": 3.209,
|
| 291 |
+
"eval_uni_ce_loss": 0.5277818014088287,
|
| 292 |
+
"step": 6144
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.2955983557341462,
|
| 296 |
+
"grad_norm": 0.8735672235488892,
|
| 297 |
+
"learning_rate": 0.0008205359904536107,
|
| 298 |
+
"loss": 1.7374849319458008,
|
| 299 |
+
"step": 6400
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.30742228996351206,
|
| 303 |
+
"grad_norm": 1.0139427185058594,
|
| 304 |
+
"learning_rate": 0.0008057078912056363,
|
| 305 |
+
"loss": 1.7405836582183838,
|
| 306 |
+
"step": 6656
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.3192462241928779,
|
| 310 |
+
"grad_norm": 1.0008057355880737,
|
| 311 |
+
"learning_rate": 0.0007904373410796086,
|
| 312 |
+
"loss": 1.7392584085464478,
|
| 313 |
+
"step": 6912
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.3310701584222438,
|
| 317 |
+
"grad_norm": 0.8208015561103821,
|
| 318 |
+
"learning_rate": 0.0007747464411350876,
|
| 319 |
+
"loss": 1.7316131591796875,
|
| 320 |
+
"step": 7168
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.3310701584222438,
|
| 324 |
+
"eval_beta_ce_loss": 0.6033149586148459,
|
| 325 |
+
"eval_bleu": 0.5017084521574802,
|
| 326 |
+
"eval_loss": 1.7307289455034962,
|
| 327 |
+
"eval_uni_ce_loss": 0.5240990275253444,
|
| 328 |
+
"step": 7168
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.3310701584222438,
|
| 332 |
+
"eval_beta_ce_loss": 0.6033149586148459,
|
| 333 |
+
"eval_bleu": 0.5017084521574802,
|
| 334 |
+
"eval_loss": 1.7307289455034962,
|
| 335 |
+
"eval_runtime": 136.4837,
|
| 336 |
+
"eval_samples_per_second": 205.101,
|
| 337 |
+
"eval_steps_per_second": 3.209,
|
| 338 |
+
"eval_uni_ce_loss": 0.5240990275253444,
|
| 339 |
+
"step": 7168
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.34289409265160964,
|
| 343 |
+
"grad_norm": 0.9193022847175598,
|
| 344 |
+
"learning_rate": 0.000758657900803716,
|
| 345 |
+
"loss": 1.7337878942489624,
|
| 346 |
+
"step": 7424
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.3547180268809755,
|
| 350 |
+
"grad_norm": 0.7078978419303894,
|
| 351 |
+
"learning_rate": 0.000742195005021869,
|
| 352 |
+
"loss": 1.7295589447021484,
|
| 353 |
+
"step": 7680
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.3665419611103413,
|
| 357 |
+
"grad_norm": 0.7894092798233032,
|
| 358 |
+
"learning_rate": 0.0007253815805303786,
|
| 359 |
+
"loss": 1.723784327507019,
|
| 360 |
+
"step": 7936
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.37836589533970716,
|
| 364 |
+
"grad_norm": 0.7812065482139587,
|
| 365 |
+
"learning_rate": 0.0007082419613901028,
|
| 366 |
+
"loss": 1.7162002325057983,
|
| 367 |
+
"step": 8192
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.37836589533970716,
|
| 371 |
+
"eval_beta_ce_loss": 0.6003280380000807,
|
| 372 |
+
"eval_bleu": 0.5041936063198118,
|
| 373 |
+
"eval_loss": 1.7231203064526597,
|
| 374 |
+
"eval_uni_ce_loss": 0.5224642294318709,
|
| 375 |
+
"step": 8192
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.37836589533970716,
|
| 379 |
+
"eval_beta_ce_loss": 0.6003280380000807,
|
| 380 |
+
"eval_bleu": 0.5041936063198118,
|
| 381 |
+
"eval_loss": 1.7231203064526597,
|
| 382 |
+
"eval_runtime": 136.1645,
|
| 383 |
+
"eval_samples_per_second": 205.582,
|
| 384 |
+
"eval_steps_per_second": 3.217,
|
| 385 |
+
"eval_uni_ce_loss": 0.5224642294318709,
|
| 386 |
+
"step": 8192
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.390189829569073,
|
| 390 |
+
"grad_norm": 0.6728302836418152,
|
| 391 |
+
"learning_rate": 0.0006908009537632514,
|
| 392 |
+
"loss": 1.7215343713760376,
|
| 393 |
+
"step": 8448
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.4020137637984389,
|
| 397 |
+
"grad_norm": 0.6643648743629456,
|
| 398 |
+
"learning_rate": 0.0006730838000114403,
|
| 399 |
+
"loss": 1.7175226211547852,
|
| 400 |
+
"step": 8704
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.41383769802780473,
|
| 404 |
+
"grad_norm": 0.6722604036331177,
|
| 405 |
+
"learning_rate": 0.0006551161421624341,
|
| 406 |
+
"loss": 1.7141364812850952,
|
| 407 |
+
"step": 8960
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.4256616322571706,
|
| 411 |
+
"grad_norm": 0.7661899328231812,
|
| 412 |
+
"learning_rate": 0.0006369239847984517,
|
| 413 |
+
"loss": 1.7067075967788696,
|
| 414 |
+
"step": 9216
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.4256616322571706,
|
| 418 |
+
"eval_beta_ce_loss": 0.5960702291906697,
|
| 419 |
+
"eval_bleu": 0.508842331362903,
|
| 420 |
+
"eval_loss": 1.7111802789718593,
|
| 421 |
+
"eval_uni_ce_loss": 0.5190398215431057,
|
| 422 |
+
"step": 9216
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.4256616322571706,
|
| 426 |
+
"eval_beta_ce_loss": 0.5960702291906697,
|
| 427 |
+
"eval_bleu": 0.508842331362903,
|
| 428 |
+
"eval_loss": 1.7111802789718593,
|
| 429 |
+
"eval_runtime": 135.4442,
|
| 430 |
+
"eval_samples_per_second": 206.676,
|
| 431 |
+
"eval_steps_per_second": 3.234,
|
| 432 |
+
"eval_uni_ce_loss": 0.5190398215431057,
|
| 433 |
+
"step": 9216
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.4374855664865364,
|
| 437 |
+
"grad_norm": 0.7040799260139465,
|
| 438 |
+
"learning_rate": 0.0006185336574197479,
|
| 439 |
+
"loss": 1.7019784450531006,
|
| 440 |
+
"step": 9472
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.44930950071590225,
|
| 444 |
+
"grad_norm": 0.7697860598564148,
|
| 445 |
+
"learning_rate": 0.0005999717763379407,
|
| 446 |
+
"loss": 1.7126665115356445,
|
| 447 |
+
"step": 9728
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.4611334349452681,
|
| 451 |
+
"grad_norm": 0.7353936433792114,
|
| 452 |
+
"learning_rate": 0.0005812652061542363,
|
| 453 |
+
"loss": 1.7056387662887573,
|
| 454 |
+
"step": 9984
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.472957369174634,
|
| 458 |
+
"grad_norm": 0.8408072590827942,
|
| 459 |
+
"learning_rate": 0.0005624410208783071,
|
| 460 |
+
"loss": 1.7016122341156006,
|
| 461 |
+
"step": 10240
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.472957369174634,
|
| 465 |
+
"eval_beta_ce_loss": 0.5948017802276567,
|
| 466 |
+
"eval_bleu": 0.5119743719435753,
|
| 467 |
+
"eval_loss": 1.705481736083009,
|
| 468 |
+
"eval_uni_ce_loss": 0.5158781752874863,
|
| 469 |
+
"step": 10240
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.472957369174634,
|
| 473 |
+
"eval_beta_ce_loss": 0.5948017802276567,
|
| 474 |
+
"eval_bleu": 0.5119743719435753,
|
| 475 |
+
"eval_loss": 1.705481736083009,
|
| 476 |
+
"eval_runtime": 134.7944,
|
| 477 |
+
"eval_samples_per_second": 207.672,
|
| 478 |
+
"eval_steps_per_second": 3.249,
|
| 479 |
+
"eval_uni_ce_loss": 0.5158781752874863,
|
| 480 |
+
"step": 10240
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.48478130340399983,
|
| 484 |
+
"grad_norm": 0.6450205445289612,
|
| 485 |
+
"learning_rate": 0.0005435264647440881,
|
| 486 |
+
"loss": 1.7052643299102783,
|
| 487 |
+
"step": 10496
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.49660523763336567,
|
| 491 |
+
"grad_norm": 0.6514373421669006,
|
| 492 |
+
"learning_rate": 0.000524548912779213,
|
| 493 |
+
"loss": 1.69474458694458,
|
| 494 |
+
"step": 10752
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.5084291718627315,
|
| 498 |
+
"grad_norm": 0.7065825462341309,
|
| 499 |
+
"learning_rate": 0.0005055358311851499,
|
| 500 |
+
"loss": 1.7007395029067993,
|
| 501 |
+
"step": 11008
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.5202531060920974,
|
| 505 |
+
"grad_norm": 0.8027070164680481,
|
| 506 |
+
"learning_rate": 0.0004865147375853812,
|
| 507 |
+
"loss": 1.696798324584961,
|
| 508 |
+
"step": 11264
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.5202531060920974,
|
| 512 |
+
"eval_beta_ce_loss": 0.5911326601625033,
|
| 513 |
+
"eval_bleu": 0.5082277536286182,
|
| 514 |
+
"eval_loss": 1.6985954277591617,
|
| 515 |
+
"eval_uni_ce_loss": 0.5163301084547827,
|
| 516 |
+
"step": 11264
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 0.5202531060920974,
|
| 520 |
+
"eval_beta_ce_loss": 0.5911326601625033,
|
| 521 |
+
"eval_bleu": 0.5082277536286182,
|
| 522 |
+
"eval_loss": 1.6985954277591617,
|
| 523 |
+
"eval_runtime": 138.4679,
|
| 524 |
+
"eval_samples_per_second": 202.162,
|
| 525 |
+
"eval_steps_per_second": 3.163,
|
| 526 |
+
"eval_uni_ce_loss": 0.5163301084547827,
|
| 527 |
+
"step": 11264
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.5320770403214632,
|
| 531 |
+
"grad_norm": 0.7604843974113464,
|
| 532 |
+
"learning_rate": 0.0004675131611991607,
|
| 533 |
+
"loss": 1.6915127038955688,
|
| 534 |
+
"step": 11520
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.5439009745508291,
|
| 538 |
+
"grad_norm": 0.6604936718940735,
|
| 539 |
+
"learning_rate": 0.0004485586029984899,
|
| 540 |
+
"loss": 1.692893147468567,
|
| 541 |
+
"step": 11776
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.5557249087801949,
|
| 545 |
+
"grad_norm": 0.7172213792800903,
|
| 546 |
+
"learning_rate": 0.00042967849590597266,
|
| 547 |
+
"loss": 1.6868789196014404,
|
| 548 |
+
"step": 12032
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.5675488430095608,
|
| 552 |
+
"grad_norm": 0.7629127502441406,
|
| 553 |
+
"learning_rate": 0.0004109001650911621,
|
| 554 |
+
"loss": 1.6939620971679688,
|
| 555 |
+
"step": 12288
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.5675488430095608,
|
| 559 |
+
"eval_beta_ce_loss": 0.5891192779965597,
|
| 560 |
+
"eval_bleu": 0.5114094116436015,
|
| 561 |
+
"eval_loss": 1.6906098777845027,
|
| 562 |
+
"eval_uni_ce_loss": 0.5123713216552995,
|
| 563 |
+
"step": 12288
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.5675488430095608,
|
| 567 |
+
"eval_beta_ce_loss": 0.5891192779965597,
|
| 568 |
+
"eval_bleu": 0.5114094116436015,
|
| 569 |
+
"eval_loss": 1.6906098777845027,
|
| 570 |
+
"eval_runtime": 135.3599,
|
| 571 |
+
"eval_samples_per_second": 206.804,
|
| 572 |
+
"eval_steps_per_second": 3.236,
|
| 573 |
+
"eval_uni_ce_loss": 0.5123713216552995,
|
| 574 |
+
"step": 12288
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 0.5793727772389267,
|
| 578 |
+
"grad_norm": 0.6822903156280518,
|
| 579 |
+
"learning_rate": 0.0003922507884228551,
|
| 580 |
+
"loss": 1.6874535083770752,
|
| 581 |
+
"step": 12544
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 0.5911967114682924,
|
| 585 |
+
"grad_norm": 0.7348518371582031,
|
| 586 |
+
"learning_rate": 0.00037375735713457723,
|
| 587 |
+
"loss": 1.6786881685256958,
|
| 588 |
+
"step": 12800
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 0.6030206456976583,
|
| 592 |
+
"grad_norm": 0.797898530960083,
|
| 593 |
+
"learning_rate": 0.00035544663676018276,
|
| 594 |
+
"loss": 1.6823772192001343,
|
| 595 |
+
"step": 13056
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 0.6148445799270241,
|
| 599 |
+
"grad_norm": 0.666439950466156,
|
| 600 |
+
"learning_rate": 0.00033734512839611255,
|
| 601 |
+
"loss": 1.6804336309432983,
|
| 602 |
+
"step": 13312
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.6148445799270241,
|
| 606 |
+
"eval_beta_ce_loss": 0.5856027225517246,
|
| 607 |
+
"eval_bleu": 0.5110656031508146,
|
| 608 |
+
"eval_loss": 1.6819111756538148,
|
| 609 |
+
"eval_uni_ce_loss": 0.5107057281689013,
|
| 610 |
+
"step": 13312
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 0.6148445799270241,
|
| 614 |
+
"eval_beta_ce_loss": 0.5856027225517246,
|
| 615 |
+
"eval_bleu": 0.5110656031508146,
|
| 616 |
+
"eval_loss": 1.6819111756538148,
|
| 617 |
+
"eval_runtime": 135.0756,
|
| 618 |
+
"eval_samples_per_second": 207.24,
|
| 619 |
+
"eval_steps_per_second": 3.243,
|
| 620 |
+
"eval_uni_ce_loss": 0.5107057281689013,
|
| 621 |
+
"step": 13312
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.62666851415639,
|
| 625 |
+
"grad_norm": 0.9536261558532715,
|
| 626 |
+
"learning_rate": 0.0003194790303463687,
|
| 627 |
+
"loss": 1.6719762086868286,
|
| 628 |
+
"step": 13568
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 0.6384924483857558,
|
| 632 |
+
"grad_norm": 0.7538688778877258,
|
| 633 |
+
"learning_rate": 0.00030187420020572406,
|
| 634 |
+
"loss": 1.6798793077468872,
|
| 635 |
+
"step": 13824
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.6503163826151217,
|
| 639 |
+
"grad_norm": 0.6934835910797119,
|
| 640 |
+
"learning_rate": 0.00028455611743603626,
|
| 641 |
+
"loss": 1.6709473133087158,
|
| 642 |
+
"step": 14080
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 0.6621403168444876,
|
| 646 |
+
"grad_norm": 0.7217169404029846,
|
| 647 |
+
"learning_rate": 0.0002675498464898373,
|
| 648 |
+
"loss": 1.6749387979507446,
|
| 649 |
+
"step": 14336
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.6621403168444876,
|
| 653 |
+
"eval_beta_ce_loss": 0.5853756803627972,
|
| 654 |
+
"eval_bleu": 0.5112546154669725,
|
| 655 |
+
"eval_loss": 1.6793161832034316,
|
| 656 |
+
"eval_uni_ce_loss": 0.5085648222737116,
|
| 657 |
+
"step": 14336
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.6621403168444876,
|
| 661 |
+
"eval_beta_ce_loss": 0.5853756803627972,
|
| 662 |
+
"eval_bleu": 0.5112546154669725,
|
| 663 |
+
"eval_loss": 1.6793161832034316,
|
| 664 |
+
"eval_runtime": 137.7212,
|
| 665 |
+
"eval_samples_per_second": 203.258,
|
| 666 |
+
"eval_steps_per_second": 3.18,
|
| 667 |
+
"eval_uni_ce_loss": 0.5085648222737116,
|
| 668 |
+
"step": 14336
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.6739642510738534,
|
| 672 |
+
"grad_norm": 0.8353447914123535,
|
| 673 |
+
"learning_rate": 0.0002508800005345623,
|
| 674 |
+
"loss": 1.679425835609436,
|
| 675 |
+
"step": 14592
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.6857881853032193,
|
| 679 |
+
"grad_norm": 0.6642665266990662,
|
| 680 |
+
"learning_rate": 0.00023457070582992562,
|
| 681 |
+
"loss": 1.6819074153900146,
|
| 682 |
+
"step": 14848
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.6976121195325851,
|
| 686 |
+
"grad_norm": 0.6584897637367249,
|
| 687 |
+
"learning_rate": 0.00021864556680999692,
|
| 688 |
+
"loss": 1.6745400428771973,
|
| 689 |
+
"step": 15104
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.709436053761951,
|
| 693 |
+
"grad_norm": 0.6884040236473083,
|
| 694 |
+
"learning_rate": 0.0002031276319205152,
|
| 695 |
+
"loss": 1.6730715036392212,
|
| 696 |
+
"step": 15360
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.709436053761951,
|
| 700 |
+
"eval_beta_ce_loss": 0.582520013870714,
|
| 701 |
+
"eval_bleu": 0.5116472036246329,
|
| 702 |
+
"eval_loss": 1.6724328953925878,
|
| 703 |
+
"eval_uni_ce_loss": 0.5073928678552854,
|
| 704 |
+
"step": 15360
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.709436053761951,
|
| 708 |
+
"eval_beta_ce_loss": 0.582520013870714,
|
| 709 |
+
"eval_bleu": 0.5116472036246329,
|
| 710 |
+
"eval_loss": 1.6724328953925878,
|
| 711 |
+
"eval_runtime": 135.5849,
|
| 712 |
+
"eval_samples_per_second": 206.461,
|
| 713 |
+
"eval_steps_per_second": 3.23,
|
| 714 |
+
"eval_uni_ce_loss": 0.5073928678552854,
|
| 715 |
+
"step": 15360
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 0.7212599879913169,
|
| 719 |
+
"grad_norm": 0.7419734597206116,
|
| 720 |
+
"learning_rate": 0.00018803936026088542,
|
| 721 |
+
"loss": 1.6732308864593506,
|
| 722 |
+
"step": 15616
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.7330839222206826,
|
| 726 |
+
"grad_norm": 0.6793375611305237,
|
| 727 |
+
"learning_rate": 0.00017340258907913464,
|
| 728 |
+
"loss": 1.6713887453079224,
|
| 729 |
+
"step": 15872
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.7449078564500485,
|
| 733 |
+
"grad_norm": 0.6444223523139954,
|
| 734 |
+
"learning_rate": 0.0001592385021668743,
|
| 735 |
+
"loss": 1.664468765258789,
|
| 736 |
+
"step": 16128
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.7567317906794143,
|
| 740 |
+
"grad_norm": 0.7465987205505371,
|
| 741 |
+
"learning_rate": 0.0001455675992000087,
|
| 742 |
+
"loss": 1.6718043088912964,
|
| 743 |
+
"step": 16384
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.7567317906794143,
|
| 747 |
+
"eval_beta_ce_loss": 0.5802708933614704,
|
| 748 |
+
"eval_bleu": 0.5137912099965332,
|
| 749 |
+
"eval_loss": 1.6668929272590707,
|
| 750 |
+
"eval_uni_ce_loss": 0.5063511407402552,
|
| 751 |
+
"step": 16384
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.7567317906794143,
|
| 755 |
+
"eval_beta_ce_loss": 0.5802708933614704,
|
| 756 |
+
"eval_bleu": 0.5137912099965332,
|
| 757 |
+
"eval_loss": 1.6668929272590707,
|
| 758 |
+
"eval_runtime": 137.3458,
|
| 759 |
+
"eval_samples_per_second": 203.814,
|
| 760 |
+
"eval_steps_per_second": 3.189,
|
| 761 |
+
"eval_uni_ce_loss": 0.5063511407402552,
|
| 762 |
+
"step": 16384
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 0.7685557249087802,
|
| 766 |
+
"grad_norm": 0.7398092746734619,
|
| 767 |
+
"learning_rate": 0.000132409666069565,
|
| 768 |
+
"loss": 1.6554030179977417,
|
| 769 |
+
"step": 16640
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 0.780379659138146,
|
| 773 |
+
"grad_norm": 0.7583303451538086,
|
| 774 |
+
"learning_rate": 0.0001197837462455823,
|
| 775 |
+
"loss": 1.6627554893493652,
|
| 776 |
+
"step": 16896
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.7922035933675119,
|
| 780 |
+
"grad_norm": 0.7260330319404602,
|
| 781 |
+
"learning_rate": 0.00010770811321550749,
|
| 782 |
+
"loss": 1.6633129119873047,
|
| 783 |
+
"step": 17152
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 0.8040275275968778,
|
| 787 |
+
"grad_norm": 0.6960119605064392,
|
| 788 |
+
"learning_rate": 9.620024403698591e-05,
|
| 789 |
+
"loss": 1.6641393899917603,
|
| 790 |
+
"step": 17408
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 0.8040275275968778,
|
| 794 |
+
"eval_beta_ce_loss": 0.580827721654008,
|
| 795 |
+
"eval_bleu": 0.5137603294282832,
|
| 796 |
+
"eval_loss": 1.6667931901809832,
|
| 797 |
+
"eval_uni_ce_loss": 0.5051377474853437,
|
| 798 |
+
"step": 17408
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 0.8040275275968778,
|
| 802 |
+
"eval_beta_ce_loss": 0.580827721654008,
|
| 803 |
+
"eval_bleu": 0.5137603294282832,
|
| 804 |
+
"eval_loss": 1.6667931901809832,
|
| 805 |
+
"eval_runtime": 135.8623,
|
| 806 |
+
"eval_samples_per_second": 206.039,
|
| 807 |
+
"eval_steps_per_second": 3.224,
|
| 808 |
+
"eval_uni_ce_loss": 0.5051377474853437,
|
| 809 |
+
"step": 17408
|
| 810 |
+
}
|
| 811 |
+
],
|
| 812 |
+
"logging_steps": 256,
|
| 813 |
+
"max_steps": 21651,
|
| 814 |
+
"num_input_tokens_seen": 0,
|
| 815 |
+
"num_train_epochs": 1,
|
| 816 |
+
"save_steps": 1024,
|
| 817 |
+
"stateful_callbacks": {
|
| 818 |
+
"TrainerControl": {
|
| 819 |
+
"args": {
|
| 820 |
+
"should_epoch_stop": false,
|
| 821 |
+
"should_evaluate": false,
|
| 822 |
+
"should_log": false,
|
| 823 |
+
"should_save": true,
|
| 824 |
+
"should_training_stop": false
|
| 825 |
+
},
|
| 826 |
+
"attributes": {}
|
| 827 |
+
}
|
| 828 |
+
},
|
| 829 |
+
"total_flos": 0.0,
|
| 830 |
+
"train_batch_size": 64,
|
| 831 |
+
"trial_name": null,
|
| 832 |
+
"trial_params": null
|
| 833 |
+
}
|
checkpoints-d2.0-t1/checkpoint-17408/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61a16bb839f687414b8e48611327c4b9cfddeefe38c031ca70808f9a97c476b7
|
| 3 |
+
size 5137
|