Training in progress, epoch 1, checkpoint
Browse files
last-checkpoint/config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"_remove_final_layer_norm": false,
|
4 |
"activation_dropout": 0.0,
|
5 |
"activation_function": "relu",
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "ccore/ccore-v3",
|
3 |
"_remove_final_layer_norm": false,
|
4 |
"activation_dropout": 0.0,
|
5 |
"activation_function": "relu",
|
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:a4013af1bfe8d529a0921d3154adc965111f5f1e71176b73e3b888780d625357
|
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 sha256:757da7d4f6896bd7517037c1352fee52b788211aeff771636b1687b889bfad65
|
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 sha256:5d4270c057ffbe7d94d3a4a46e15a61be9cfe5f3368f1f48731c220c26ede867
|
3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,113 +1,24 @@
|
|
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": 0
|
13 |
-
"
|
14 |
-
"
|
15 |
-
"
|
16 |
-
"
|
17 |
-
|
18 |
-
{
|
19 |
-
"epoch": 0.998769987699877,
|
20 |
-
"eval_loss": 1.013350248336792,
|
21 |
-
"eval_runtime": 259.9403,
|
22 |
-
"eval_samples_per_second": 50.035,
|
23 |
-
"eval_steps_per_second": 12.511,
|
24 |
-
"step": 609
|
25 |
-
},
|
26 |
-
{
|
27 |
-
"epoch": 1.6412464124641246,
|
28 |
-
"grad_norm": 2.491156578063965,
|
29 |
-
"learning_rate": 9.351884318196129e-05,
|
30 |
-
"loss": 1.0221,
|
31 |
-
"step": 1000
|
32 |
-
},
|
33 |
-
{
|
34 |
-
"epoch": 1.998769987699877,
|
35 |
-
"eval_loss": 1.0091875791549683,
|
36 |
-
"eval_runtime": 259.5356,
|
37 |
-
"eval_samples_per_second": 50.113,
|
38 |
-
"eval_steps_per_second": 12.53,
|
39 |
-
"step": 1218
|
40 |
-
},
|
41 |
-
{
|
42 |
-
"epoch": 2.4624846248462484,
|
43 |
-
"grad_norm": 2.6502525806427,
|
44 |
-
"learning_rate": 8.579934515124202e-05,
|
45 |
-
"loss": 1.0175,
|
46 |
-
"step": 1500
|
47 |
-
},
|
48 |
-
{
|
49 |
-
"epoch": 2.998769987699877,
|
50 |
-
"eval_loss": 1.0066354274749756,
|
51 |
-
"eval_runtime": 262.6041,
|
52 |
-
"eval_samples_per_second": 49.527,
|
53 |
-
"eval_steps_per_second": 12.384,
|
54 |
-
"step": 1827
|
55 |
-
},
|
56 |
-
{
|
57 |
-
"epoch": 3.2837228372283724,
|
58 |
-
"grad_norm": 1.8426792621612549,
|
59 |
-
"learning_rate": 7.571135915407219e-05,
|
60 |
-
"loss": 1.0148,
|
61 |
-
"step": 2000
|
62 |
-
},
|
63 |
-
{
|
64 |
-
"epoch": 3.998769987699877,
|
65 |
-
"eval_loss": 1.003442406654358,
|
66 |
-
"eval_runtime": 261.9157,
|
67 |
-
"eval_samples_per_second": 49.657,
|
68 |
-
"eval_steps_per_second": 12.416,
|
69 |
-
"step": 2436
|
70 |
-
},
|
71 |
-
{
|
72 |
-
"epoch": 4.1049610496104965,
|
73 |
-
"grad_norm": 1.7222294807434082,
|
74 |
-
"learning_rate": 6.39223072714725e-05,
|
75 |
-
"loss": 1.0105,
|
76 |
-
"step": 2500
|
77 |
-
},
|
78 |
-
{
|
79 |
-
"epoch": 4.924969249692497,
|
80 |
-
"grad_norm": 3.702568292617798,
|
81 |
-
"learning_rate": 5.121215425983256e-05,
|
82 |
-
"loss": 1.0048,
|
83 |
-
"step": 3000
|
84 |
-
},
|
85 |
-
{
|
86 |
-
"epoch": 4.998769987699877,
|
87 |
-
"eval_loss": 0.9986960887908936,
|
88 |
-
"eval_runtime": 258.2811,
|
89 |
-
"eval_samples_per_second": 50.356,
|
90 |
-
"eval_steps_per_second": 12.591,
|
91 |
-
"step": 3045
|
92 |
-
},
|
93 |
-
{
|
94 |
-
"epoch": 5.74620746207462,
|
95 |
-
"grad_norm": 1.9525970220565796,
|
96 |
-
"learning_rate": 3.842180501092163e-05,
|
97 |
-
"loss": 1.0013,
|
98 |
-
"step": 3500
|
99 |
-
},
|
100 |
-
{
|
101 |
-
"epoch": 5.998769987699877,
|
102 |
-
"eval_loss": 0.9929500818252563,
|
103 |
-
"eval_runtime": 260.4505,
|
104 |
-
"eval_samples_per_second": 49.937,
|
105 |
-
"eval_steps_per_second": 12.486,
|
106 |
-
"step": 3654
|
107 |
}
|
108 |
],
|
109 |
"logging_steps": 500,
|
110 |
-
"max_steps":
|
111 |
"num_input_tokens_seen": 0,
|
112 |
"num_train_epochs": 10,
|
113 |
"save_steps": 500,
|
@@ -123,7 +34,7 @@
|
|
123 |
"attributes": {}
|
124 |
}
|
125 |
},
|
126 |
-
"total_flos":
|
127 |
"train_batch_size": 24,
|
128 |
"trial_name": null,
|
129 |
"trial_params": null
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.7184422016143799,
|
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": 0.7184422016143799,
|
14 |
+
"eval_runtime": 5.9548,
|
15 |
+
"eval_samples_per_second": 50.38,
|
16 |
+
"eval_steps_per_second": 12.595,
|
17 |
+
"step": 15
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
}
|
19 |
],
|
20 |
"logging_steps": 500,
|
21 |
+
"max_steps": 140,
|
22 |
"num_input_tokens_seen": 0,
|
23 |
"num_train_epochs": 10,
|
24 |
"save_steps": 500,
|
|
|
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:ce5b37466798b22cd1c58fa164a2deb26fd86f95d9733731d6588e8abd389d38
|
3 |
size 5368
|