Training in progress, step 5
Browse files- last-checkpoint/optimizer.pt +1 -1
- last-checkpoint/pytorch_model.bin +1 -1
- last-checkpoint/rng_state.pth +1 -1
- last-checkpoint/scaler.pt +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +27 -3
- pytorch_model.bin +1 -1
- runs/Jul19_18-03-38_47ad500bc002/events.out.tfevents.1689789837.47ad500bc002.2216.8 +2 -2
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1044539653
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0c32be757f0288a7c67461a859c5f4d72e961a7d5c562d0fa1664c73de6c8154
|
3 |
size 1044539653
|
last-checkpoint/pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5018531153
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:816a60f80d1cb9952d19b233fc387797608aafd271895e74c640a0a12d4a308e
|
3 |
size 5018531153
|
last-checkpoint/rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14575
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9ad35f6fad5899e2120c58dd6f8e19da6576502e2471f7fbb0d8f573b5b8c4cd
|
3 |
size 14575
|
last-checkpoint/scaler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 557
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4226304fb62e961f1febc70be2db3d4aa0fa092f6ffb6147df348b679da41079
|
3 |
size 557
|
last-checkpoint/scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 627
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bfad7054a4f055c4b3d58bee7c17c677cfc1a348a8f81a9063089f8cb55dc31e
|
3 |
size 627
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -12,11 +12,35 @@
|
|
12 |
"learning_rate": 0.0002,
|
13 |
"loss": 1.7413,
|
14 |
"step": 1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
}
|
16 |
],
|
17 |
"max_steps": 500,
|
18 |
"num_train_epochs": 125,
|
19 |
-
"total_flos":
|
20 |
"trial_name": null,
|
21 |
"trial_params": null
|
22 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.25,
|
5 |
+
"global_step": 5,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
12 |
"learning_rate": 0.0002,
|
13 |
"loss": 1.7413,
|
14 |
"step": 1
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.5,
|
18 |
+
"learning_rate": 0.0002,
|
19 |
+
"loss": 1.7187,
|
20 |
+
"step": 2
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.75,
|
24 |
+
"learning_rate": 0.0002,
|
25 |
+
"loss": 1.7006,
|
26 |
+
"step": 3
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 1.0,
|
30 |
+
"learning_rate": 0.0002,
|
31 |
+
"loss": 1.5806,
|
32 |
+
"step": 4
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.25,
|
36 |
+
"learning_rate": 0.0002,
|
37 |
+
"loss": 1.5316,
|
38 |
+
"step": 5
|
39 |
}
|
40 |
],
|
41 |
"max_steps": 500,
|
42 |
"num_train_epochs": 125,
|
43 |
+
"total_flos": 835774634655744.0,
|
44 |
"trial_name": null,
|
45 |
"trial_params": null
|
46 |
}
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5018531153
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:816a60f80d1cb9952d19b233fc387797608aafd271895e74c640a0a12d4a308e
|
3 |
size 5018531153
|
runs/Jul19_18-03-38_47ad500bc002/events.out.tfevents.1689789837.47ad500bc002.2216.8
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 sha256:3a1ff8931a8754e3d54dd36950dce6cd0ee70d0b9b8e7bd51c1948bdf67c08b2
|
3 |
+
size 5290
|