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 557116312
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0321845aa554c15389dd24dff69e92c7373fc5b04cb68327a41cdc98bbcd31ba
|
3 |
size 557116312
|
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1113991930
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:62afe7d04b548ec8eec3d1689d7d8c3e8d969ad62a70b280216e862ba19c8547
|
3 |
size 1113991930
|
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 sha256:59624509e89bbc9b94c591ed4735982e33caa7e6875690c4ff16d1a97717d87c
|
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 sha256:9ac9643ef164647dd74d17c68b79347a9f51627588cd7ec3dd1a18bf83f7e68f
|
3 |
size 1064
|
last-checkpoint/tokenizer.json
CHANGED
@@ -1,6 +1,11 @@
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
-
"truncation":
|
|
|
|
|
|
|
|
|
|
|
4 |
"padding": null,
|
5 |
"added_tokens": [
|
6 |
{
|
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
+
"truncation": {
|
4 |
+
"direction": "Right",
|
5 |
+
"max_length": 128,
|
6 |
+
"strategy": "LongestFirst",
|
7 |
+
"stride": 0
|
8 |
+
},
|
9 |
"padding": null,
|
10 |
"added_tokens": [
|
11 |
{
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,146 +1,50 @@
|
|
1 |
{
|
2 |
-
"best_metric":
|
3 |
-
"best_model_checkpoint": "my-model/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": 0.32722513089005234,
|
13 |
-
"grad_norm":
|
14 |
-
"learning_rate": 4.
|
15 |
-
"loss": 0.
|
16 |
"step": 500
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.6544502617801047,
|
20 |
-
"grad_norm":
|
21 |
-
"learning_rate":
|
22 |
-
"loss": 0.
|
23 |
"step": 1000
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.981675392670157,
|
27 |
-
"grad_norm": 1.
|
28 |
-
"learning_rate": 3.
|
29 |
-
"loss": 0.
|
30 |
"step": 1500
|
31 |
},
|
32 |
{
|
33 |
"epoch": 1.0,
|
34 |
-
"eval_loss": 5.
|
35 |
-
"eval_rouge1":
|
36 |
-
"eval_rouge2":
|
37 |
-
"eval_rougeL": 21.
|
38 |
-
"eval_runtime":
|
39 |
-
"eval_samples_per_second": 5.
|
40 |
-
"eval_steps_per_second": 0.
|
41 |
"step": 1528
|
42 |
-
},
|
43 |
-
{
|
44 |
-
"epoch": 1.3089005235602094,
|
45 |
-
"grad_norm": 3.42366361618042,
|
46 |
-
"learning_rate": 3.3638743455497386e-05,
|
47 |
-
"loss": 0.0539,
|
48 |
-
"step": 2000
|
49 |
-
},
|
50 |
-
{
|
51 |
-
"epoch": 1.6361256544502618,
|
52 |
-
"grad_norm": 3.776289701461792,
|
53 |
-
"learning_rate": 2.954842931937173e-05,
|
54 |
-
"loss": 0.0727,
|
55 |
-
"step": 2500
|
56 |
-
},
|
57 |
-
{
|
58 |
-
"epoch": 1.9633507853403143,
|
59 |
-
"grad_norm": 5.072182655334473,
|
60 |
-
"learning_rate": 2.545811518324607e-05,
|
61 |
-
"loss": 0.1352,
|
62 |
-
"step": 3000
|
63 |
-
},
|
64 |
-
{
|
65 |
-
"epoch": 2.0,
|
66 |
-
"eval_loss": 5.219298839569092,
|
67 |
-
"eval_rouge1": 24.8743,
|
68 |
-
"eval_rouge2": 9.8439,
|
69 |
-
"eval_rougeL": 21.6411,
|
70 |
-
"eval_runtime": 433.0859,
|
71 |
-
"eval_samples_per_second": 5.068,
|
72 |
-
"eval_steps_per_second": 0.508,
|
73 |
-
"step": 3056
|
74 |
-
},
|
75 |
-
{
|
76 |
-
"epoch": 2.2905759162303667,
|
77 |
-
"grad_norm": 6.1271209716796875,
|
78 |
-
"learning_rate": 2.136780104712042e-05,
|
79 |
-
"loss": 0.4727,
|
80 |
-
"step": 3500
|
81 |
-
},
|
82 |
-
{
|
83 |
-
"epoch": 2.6178010471204187,
|
84 |
-
"grad_norm": 5.5732316970825195,
|
85 |
-
"learning_rate": 1.7277486910994763e-05,
|
86 |
-
"loss": 0.6959,
|
87 |
-
"step": 4000
|
88 |
-
},
|
89 |
-
{
|
90 |
-
"epoch": 2.945026178010471,
|
91 |
-
"grad_norm": 5.73837423324585,
|
92 |
-
"learning_rate": 1.3187172774869111e-05,
|
93 |
-
"loss": 0.9382,
|
94 |
-
"step": 4500
|
95 |
-
},
|
96 |
-
{
|
97 |
-
"epoch": 3.0,
|
98 |
-
"eval_loss": 3.44022798538208,
|
99 |
-
"eval_rouge1": 26.0341,
|
100 |
-
"eval_rouge2": 10.6222,
|
101 |
-
"eval_rougeL": 22.7685,
|
102 |
-
"eval_runtime": 417.9997,
|
103 |
-
"eval_samples_per_second": 5.251,
|
104 |
-
"eval_steps_per_second": 0.526,
|
105 |
-
"step": 4584
|
106 |
-
},
|
107 |
-
{
|
108 |
-
"epoch": 3.2722513089005236,
|
109 |
-
"grad_norm": 5.353453159332275,
|
110 |
-
"learning_rate": 9.096858638743457e-06,
|
111 |
-
"loss": 0.9969,
|
112 |
-
"step": 5000
|
113 |
-
},
|
114 |
-
{
|
115 |
-
"epoch": 3.599476439790576,
|
116 |
-
"grad_norm": 5.323329925537109,
|
117 |
-
"learning_rate": 5.006544502617801e-06,
|
118 |
-
"loss": 1.0924,
|
119 |
-
"step": 5500
|
120 |
-
},
|
121 |
-
{
|
122 |
-
"epoch": 3.9267015706806285,
|
123 |
-
"grad_norm": 6.1474175453186035,
|
124 |
-
"learning_rate": 9.162303664921465e-07,
|
125 |
-
"loss": 1.2208,
|
126 |
-
"step": 6000
|
127 |
-
},
|
128 |
-
{
|
129 |
-
"epoch": 4.0,
|
130 |
-
"eval_loss": 3.21282958984375,
|
131 |
-
"eval_rouge1": 26.4112,
|
132 |
-
"eval_rouge2": 10.9605,
|
133 |
-
"eval_rougeL": 23.0258,
|
134 |
-
"eval_runtime": 417.2936,
|
135 |
-
"eval_samples_per_second": 5.26,
|
136 |
-
"eval_steps_per_second": 0.527,
|
137 |
-
"step": 6112
|
138 |
}
|
139 |
],
|
140 |
"logging_steps": 500,
|
141 |
-
"max_steps":
|
142 |
"num_input_tokens_seen": 0,
|
143 |
-
"num_train_epochs":
|
144 |
"save_steps": 500,
|
145 |
"stateful_callbacks": {
|
146 |
"TrainerControl": {
|
@@ -149,12 +53,12 @@
|
|
149 |
"should_evaluate": false,
|
150 |
"should_log": false,
|
151 |
"should_save": true,
|
152 |
-
"should_training_stop":
|
153 |
},
|
154 |
"attributes": {}
|
155 |
}
|
156 |
},
|
157 |
-
"total_flos":
|
158 |
"train_batch_size": 10,
|
159 |
"trial_name": null,
|
160 |
"trial_params": null
|
|
|
1 |
{
|
2 |
+
"best_metric": 25.1709,
|
3 |
+
"best_model_checkpoint": "my-model/checkpoint-1528",
|
4 |
+
"epoch": 1.0,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 1528,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.32722513089005234,
|
13 |
+
"grad_norm": 0.9616150856018066,
|
14 |
+
"learning_rate": 4.454624781849913e-05,
|
15 |
+
"loss": 0.0158,
|
16 |
"step": 500
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.6544502617801047,
|
20 |
+
"grad_norm": 2.0808584690093994,
|
21 |
+
"learning_rate": 3.909249563699826e-05,
|
22 |
+
"loss": 0.0223,
|
23 |
"step": 1000
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.981675392670157,
|
27 |
+
"grad_norm": 1.145504117012024,
|
28 |
+
"learning_rate": 3.3638743455497386e-05,
|
29 |
+
"loss": 0.0271,
|
30 |
"step": 1500
|
31 |
},
|
32 |
{
|
33 |
"epoch": 1.0,
|
34 |
+
"eval_loss": 5.483295917510986,
|
35 |
+
"eval_rouge1": 25.1709,
|
36 |
+
"eval_rouge2": 10.0024,
|
37 |
+
"eval_rougeL": 21.9926,
|
38 |
+
"eval_runtime": 372.8877,
|
39 |
+
"eval_samples_per_second": 5.886,
|
40 |
+
"eval_steps_per_second": 0.59,
|
41 |
"step": 1528
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
42 |
}
|
43 |
],
|
44 |
"logging_steps": 500,
|
45 |
+
"max_steps": 4584,
|
46 |
"num_input_tokens_seen": 0,
|
47 |
+
"num_train_epochs": 3,
|
48 |
"save_steps": 500,
|
49 |
"stateful_callbacks": {
|
50 |
"TrainerControl": {
|
|
|
53 |
"should_evaluate": false,
|
54 |
"should_log": false,
|
55 |
"should_save": true,
|
56 |
+
"should_training_stop": false
|
57 |
},
|
58 |
"attributes": {}
|
59 |
}
|
60 |
},
|
61 |
+
"total_flos": 7327420304523264.0,
|
62 |
"train_batch_size": 10,
|
63 |
"trial_name": null,
|
64 |
"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 6968
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9454715524d57fa2a7a3e1d17cb27d9ba1a09c66d019637bdd23107b92238a57
|
3 |
size 6968
|