Training in progress, epoch 1, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 500979600
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:77a43afa5413209488282deaba8f3e6de65e3901e35bdafd4f239b63a96e4c53
|
3 |
size 500979600
|
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1002078330
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cf433d541d14e0716f77c5d6992896aed2f222236bee78e7fc2cb9d99f58d9e7
|
3 |
size 1002078330
|
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 sha256:574024fd0bef58b4ca8af0606c1be9cf07d7494e9e30913df874b1f6896f01c2
|
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 sha256:902ae1f31337b1589c09fb3fa5e1c9d37d43036b50a2a55706a5b53db24b476c
|
3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,92 +1,20 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "./opt_trained3/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": 1.0,
|
13 |
-
"eval_loss": 0.
|
14 |
-
"eval_runtime":
|
15 |
-
"eval_samples_per_second":
|
16 |
-
"eval_steps_per_second": 12.
|
17 |
"step": 15
|
18 |
-
},
|
19 |
-
{
|
20 |
-
"epoch": 2.0,
|
21 |
-
"eval_loss": 0.5414664149284363,
|
22 |
-
"eval_runtime": 5.9806,
|
23 |
-
"eval_samples_per_second": 50.162,
|
24 |
-
"eval_steps_per_second": 12.541,
|
25 |
-
"step": 30
|
26 |
-
},
|
27 |
-
{
|
28 |
-
"epoch": 3.0,
|
29 |
-
"eval_loss": 0.5381864309310913,
|
30 |
-
"eval_runtime": 5.9811,
|
31 |
-
"eval_samples_per_second": 50.158,
|
32 |
-
"eval_steps_per_second": 12.539,
|
33 |
-
"step": 45
|
34 |
-
},
|
35 |
-
{
|
36 |
-
"epoch": 4.0,
|
37 |
-
"eval_loss": 0.5374372005462646,
|
38 |
-
"eval_runtime": 5.9965,
|
39 |
-
"eval_samples_per_second": 50.03,
|
40 |
-
"eval_steps_per_second": 12.507,
|
41 |
-
"step": 60
|
42 |
-
},
|
43 |
-
{
|
44 |
-
"epoch": 5.0,
|
45 |
-
"eval_loss": 0.5423746705055237,
|
46 |
-
"eval_runtime": 5.9878,
|
47 |
-
"eval_samples_per_second": 50.102,
|
48 |
-
"eval_steps_per_second": 12.525,
|
49 |
-
"step": 75
|
50 |
-
},
|
51 |
-
{
|
52 |
-
"epoch": 6.0,
|
53 |
-
"eval_loss": 0.5467855930328369,
|
54 |
-
"eval_runtime": 5.9773,
|
55 |
-
"eval_samples_per_second": 50.19,
|
56 |
-
"eval_steps_per_second": 12.547,
|
57 |
-
"step": 90
|
58 |
-
},
|
59 |
-
{
|
60 |
-
"epoch": 7.0,
|
61 |
-
"eval_loss": 0.5484521389007568,
|
62 |
-
"eval_runtime": 5.9799,
|
63 |
-
"eval_samples_per_second": 50.168,
|
64 |
-
"eval_steps_per_second": 12.542,
|
65 |
-
"step": 105
|
66 |
-
},
|
67 |
-
{
|
68 |
-
"epoch": 8.0,
|
69 |
-
"eval_loss": 0.548169732093811,
|
70 |
-
"eval_runtime": 5.9928,
|
71 |
-
"eval_samples_per_second": 50.06,
|
72 |
-
"eval_steps_per_second": 12.515,
|
73 |
-
"step": 120
|
74 |
-
},
|
75 |
-
{
|
76 |
-
"epoch": 9.0,
|
77 |
-
"eval_loss": 0.5496641993522644,
|
78 |
-
"eval_runtime": 5.9832,
|
79 |
-
"eval_samples_per_second": 50.141,
|
80 |
-
"eval_steps_per_second": 12.535,
|
81 |
-
"step": 135
|
82 |
-
},
|
83 |
-
{
|
84 |
-
"epoch": 9.353982300884956,
|
85 |
-
"eval_loss": 0.5497786998748779,
|
86 |
-
"eval_runtime": 6.0047,
|
87 |
-
"eval_samples_per_second": 49.961,
|
88 |
-
"eval_steps_per_second": 12.49,
|
89 |
-
"step": 140
|
90 |
}
|
91 |
],
|
92 |
"logging_steps": 500,
|
@@ -101,12 +29,12 @@
|
|
101 |
"should_evaluate": false,
|
102 |
"should_log": false,
|
103 |
"should_save": true,
|
104 |
-
"should_training_stop":
|
105 |
},
|
106 |
"attributes": {}
|
107 |
}
|
108 |
},
|
109 |
-
"total_flos":
|
110 |
"train_batch_size": 24,
|
111 |
"trial_name": null,
|
112 |
"trial_params": null
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.535097062587738,
|
3 |
+
"best_model_checkpoint": "./opt_trained3/checkpoint-15",
|
4 |
+
"epoch": 1.0,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 15,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
+
"eval_loss": 0.535097062587738,
|
14 |
+
"eval_runtime": 6.0207,
|
15 |
+
"eval_samples_per_second": 49.828,
|
16 |
+
"eval_steps_per_second": 12.457,
|
17 |
"step": 15
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
}
|
19 |
],
|
20 |
"logging_steps": 500,
|
|
|
29 |
"should_evaluate": false,
|
30 |
"should_log": false,
|
31 |
"should_save": true,
|
32 |
+
"should_training_stop": false
|
33 |
},
|
34 |
"attributes": {}
|
35 |
}
|
36 |
},
|
37 |
+
"total_flos": 556605613440000.0,
|
38 |
"train_batch_size": 24,
|
39 |
"trial_name": null,
|
40 |
"trial_params": null
|
last-checkpoint/training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5368
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:da1113fb33b5dc52d2cd8aa9933b02ed905b55e4f161767b2b195fd21d946a9d
|
3 |
size 5368
|