Training in progress, step 12, checkpoint
Browse files
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 19859140
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a3c90e5845f49a2c463e517e78aef705f9618fdfc9c7660f0b3344d3608e5504
|
3 |
size 19859140
|
last-checkpoint/rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14244
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:65feb82aeff84f74730d241f8a93f4edc4ac56bf11dc7a0aebce45c910102b64
|
3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1064
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e59b38b7699f7635523e7ab6d478b4e9b420f110d0bbbef9f4bd629aa3d6cd28
|
3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch": 0.
|
5 |
"eval_steps": 3,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
@@ -31,6 +31,29 @@
|
|
31 |
"eval_samples_per_second": 52.707,
|
32 |
"eval_steps_per_second": 6.658,
|
33 |
"step": 6
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
34 |
}
|
35 |
],
|
36 |
"logging_steps": 10,
|
@@ -50,7 +73,7 @@
|
|
50 |
"attributes": {}
|
51 |
}
|
52 |
},
|
53 |
-
"total_flos":
|
54 |
"train_batch_size": 8,
|
55 |
"trial_name": null,
|
56 |
"trial_params": null
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.8495575221238938,
|
5 |
"eval_steps": 3,
|
6 |
+
"global_step": 12,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
|
|
31 |
"eval_samples_per_second": 52.707,
|
32 |
"eval_steps_per_second": 6.658,
|
33 |
"step": 6
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.6371681415929203,
|
37 |
+
"eval_loss": NaN,
|
38 |
+
"eval_runtime": 1.7975,
|
39 |
+
"eval_samples_per_second": 52.85,
|
40 |
+
"eval_steps_per_second": 6.676,
|
41 |
+
"step": 9
|
42 |
+
},
|
43 |
+
{
|
44 |
+
"epoch": 0.7079646017699115,
|
45 |
+
"grad_norm": NaN,
|
46 |
+
"learning_rate": 3.226975564787322e-05,
|
47 |
+
"loss": 0.0,
|
48 |
+
"step": 10
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 0.8495575221238938,
|
52 |
+
"eval_loss": NaN,
|
53 |
+
"eval_runtime": 1.798,
|
54 |
+
"eval_samples_per_second": 52.837,
|
55 |
+
"eval_steps_per_second": 6.674,
|
56 |
+
"step": 12
|
57 |
}
|
58 |
],
|
59 |
"logging_steps": 10,
|
|
|
73 |
"attributes": {}
|
74 |
}
|
75 |
},
|
76 |
+
"total_flos": 6226526563467264.0,
|
77 |
"train_batch_size": 8,
|
78 |
"trial_name": null,
|
79 |
"trial_params": null
|