Model save
Browse files- README.md +65 -0
- all_results.json +9 -0
- runs/May30_03-56-39_COE-CS-sv003/events.out.tfevents.1717041798.COE-CS-sv003.3598473.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +624 -0
README.md
ADDED
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: llama3
|
3 |
+
base_model: meta-llama/Meta-Llama-3-8B
|
4 |
+
tags:
|
5 |
+
- trl
|
6 |
+
- sft
|
7 |
+
- generated_from_trainer
|
8 |
+
model-index:
|
9 |
+
- name: llama20pEnt
|
10 |
+
results: []
|
11 |
+
---
|
12 |
+
|
13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
+
should probably proofread and complete it, then remove this comment. -->
|
15 |
+
|
16 |
+
# llama20pEnt
|
17 |
+
|
18 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the None dataset.
|
19 |
+
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 1.0929
|
21 |
+
|
22 |
+
## Model description
|
23 |
+
|
24 |
+
More information needed
|
25 |
+
|
26 |
+
## Intended uses & limitations
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Training and evaluation data
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Training procedure
|
35 |
+
|
36 |
+
### Training hyperparameters
|
37 |
+
|
38 |
+
The following hyperparameters were used during training:
|
39 |
+
- learning_rate: 0.0002
|
40 |
+
- train_batch_size: 8
|
41 |
+
- eval_batch_size: 4
|
42 |
+
- seed: 42
|
43 |
+
- distributed_type: multi-GPU
|
44 |
+
- num_devices: 4
|
45 |
+
- gradient_accumulation_steps: 8
|
46 |
+
- total_train_batch_size: 256
|
47 |
+
- total_eval_batch_size: 16
|
48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
49 |
+
- lr_scheduler_type: cosine
|
50 |
+
- lr_scheduler_warmup_ratio: 0.1
|
51 |
+
- num_epochs: 1
|
52 |
+
|
53 |
+
### Training results
|
54 |
+
|
55 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
56 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
57 |
+
| 0.8215 | 1.0 | 406 | 1.0929 |
|
58 |
+
|
59 |
+
|
60 |
+
### Framework versions
|
61 |
+
|
62 |
+
- Transformers 4.41.1
|
63 |
+
- Pytorch 2.3.0+cu121
|
64 |
+
- Datasets 2.19.1
|
65 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"total_flos": 1.9785163757689242e+18,
|
4 |
+
"train_loss": 0.8540300870763844,
|
5 |
+
"train_runtime": 31542.2056,
|
6 |
+
"train_samples": 103932,
|
7 |
+
"train_samples_per_second": 3.295,
|
8 |
+
"train_steps_per_second": 0.013
|
9 |
+
}
|
runs/May30_03-56-39_COE-CS-sv003/events.out.tfevents.1717041798.COE-CS-sv003.3598473.0
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:e48aa57fc1421b28f02914a3c28c2e63a936cea4aa065a5c76dfd6590fda24cf
|
3 |
+
size 22887
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"total_flos": 1.9785163757689242e+18,
|
4 |
+
"train_loss": 0.8540300870763844,
|
5 |
+
"train_runtime": 31542.2056,
|
6 |
+
"train_samples": 103932,
|
7 |
+
"train_samples_per_second": 3.295,
|
8 |
+
"train_steps_per_second": 0.013
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,624 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 406,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0024630541871921183,
|
13 |
+
"grad_norm": 0.32156336307525635,
|
14 |
+
"learning_rate": 4.8780487804878055e-06,
|
15 |
+
"loss": 1.0055,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.012315270935960592,
|
20 |
+
"grad_norm": 0.3402904272079468,
|
21 |
+
"learning_rate": 2.4390243902439026e-05,
|
22 |
+
"loss": 0.9699,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.024630541871921183,
|
27 |
+
"grad_norm": 0.24658028781414032,
|
28 |
+
"learning_rate": 4.878048780487805e-05,
|
29 |
+
"loss": 1.0011,
|
30 |
+
"step": 10
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.03694581280788178,
|
34 |
+
"grad_norm": 0.11734232306480408,
|
35 |
+
"learning_rate": 7.317073170731707e-05,
|
36 |
+
"loss": 0.9502,
|
37 |
+
"step": 15
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.04926108374384237,
|
41 |
+
"grad_norm": 0.09808443486690521,
|
42 |
+
"learning_rate": 9.75609756097561e-05,
|
43 |
+
"loss": 0.9388,
|
44 |
+
"step": 20
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.06157635467980296,
|
48 |
+
"grad_norm": 0.0954531654715538,
|
49 |
+
"learning_rate": 0.00012195121951219512,
|
50 |
+
"loss": 0.8962,
|
51 |
+
"step": 25
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.07389162561576355,
|
55 |
+
"grad_norm": 0.08142588287591934,
|
56 |
+
"learning_rate": 0.00014634146341463414,
|
57 |
+
"loss": 0.8666,
|
58 |
+
"step": 30
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.08620689655172414,
|
62 |
+
"grad_norm": 0.07427375018596649,
|
63 |
+
"learning_rate": 0.0001707317073170732,
|
64 |
+
"loss": 0.8724,
|
65 |
+
"step": 35
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.09852216748768473,
|
69 |
+
"grad_norm": 0.07633435726165771,
|
70 |
+
"learning_rate": 0.0001951219512195122,
|
71 |
+
"loss": 0.8744,
|
72 |
+
"step": 40
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.11083743842364532,
|
76 |
+
"grad_norm": 0.06420885771512985,
|
77 |
+
"learning_rate": 0.0001999407400739705,
|
78 |
+
"loss": 0.8774,
|
79 |
+
"step": 45
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.12315270935960591,
|
83 |
+
"grad_norm": 0.06878102570772171,
|
84 |
+
"learning_rate": 0.00019970011699250152,
|
85 |
+
"loss": 0.8515,
|
86 |
+
"step": 50
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.1354679802955665,
|
90 |
+
"grad_norm": 0.06849724054336548,
|
91 |
+
"learning_rate": 0.00019927487224577402,
|
92 |
+
"loss": 0.8727,
|
93 |
+
"step": 55
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.1477832512315271,
|
97 |
+
"grad_norm": 0.06974564492702484,
|
98 |
+
"learning_rate": 0.0001986657932891657,
|
99 |
+
"loss": 0.8476,
|
100 |
+
"step": 60
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.16009852216748768,
|
104 |
+
"grad_norm": 0.07568373531103134,
|
105 |
+
"learning_rate": 0.00019787400799669154,
|
106 |
+
"loss": 0.8703,
|
107 |
+
"step": 65
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.1724137931034483,
|
111 |
+
"grad_norm": 0.06864025443792343,
|
112 |
+
"learning_rate": 0.00019690098257244064,
|
113 |
+
"loss": 0.8473,
|
114 |
+
"step": 70
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.18472906403940886,
|
118 |
+
"grad_norm": 0.075719453394413,
|
119 |
+
"learning_rate": 0.00019574851883550395,
|
120 |
+
"loss": 0.8635,
|
121 |
+
"step": 75
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.19704433497536947,
|
125 |
+
"grad_norm": 0.0714876800775528,
|
126 |
+
"learning_rate": 0.00019441875088341997,
|
127 |
+
"loss": 0.8644,
|
128 |
+
"step": 80
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.20935960591133004,
|
132 |
+
"grad_norm": 0.07155413180589676,
|
133 |
+
"learning_rate": 0.00019291414114031743,
|
134 |
+
"loss": 0.8585,
|
135 |
+
"step": 85
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.22167487684729065,
|
139 |
+
"grad_norm": 0.07434839755296707,
|
140 |
+
"learning_rate": 0.00019123747579707275,
|
141 |
+
"loss": 0.8648,
|
142 |
+
"step": 90
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.23399014778325122,
|
146 |
+
"grad_norm": 0.07367011904716492,
|
147 |
+
"learning_rate": 0.0001893918596519257,
|
148 |
+
"loss": 0.8375,
|
149 |
+
"step": 95
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.24630541871921183,
|
153 |
+
"grad_norm": 0.08377306163311005,
|
154 |
+
"learning_rate": 0.00018738071036110808,
|
155 |
+
"loss": 0.8306,
|
156 |
+
"step": 100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.25862068965517243,
|
160 |
+
"grad_norm": 0.07255107164382935,
|
161 |
+
"learning_rate": 0.00018520775211013093,
|
162 |
+
"loss": 0.8511,
|
163 |
+
"step": 105
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.270935960591133,
|
167 |
+
"grad_norm": 0.08693832159042358,
|
168 |
+
"learning_rate": 0.00018287700871745036,
|
169 |
+
"loss": 0.8445,
|
170 |
+
"step": 110
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.2832512315270936,
|
174 |
+
"grad_norm": 0.08710788935422897,
|
175 |
+
"learning_rate": 0.00018039279618328212,
|
176 |
+
"loss": 0.8752,
|
177 |
+
"step": 115
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.2955665024630542,
|
181 |
+
"grad_norm": 0.09619166702032089,
|
182 |
+
"learning_rate": 0.0001777597146973627,
|
183 |
+
"loss": 0.8671,
|
184 |
+
"step": 120
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.3078817733990148,
|
188 |
+
"grad_norm": 0.07760035991668701,
|
189 |
+
"learning_rate": 0.00017498264012045687,
|
190 |
+
"loss": 0.8559,
|
191 |
+
"step": 125
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.32019704433497537,
|
195 |
+
"grad_norm": 0.0805315300822258,
|
196 |
+
"learning_rate": 0.00017206671495538612,
|
197 |
+
"loss": 0.8585,
|
198 |
+
"step": 130
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.33251231527093594,
|
202 |
+
"grad_norm": 0.07385493814945221,
|
203 |
+
"learning_rate": 0.0001690173388242972,
|
204 |
+
"loss": 0.8467,
|
205 |
+
"step": 135
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.3448275862068966,
|
209 |
+
"grad_norm": 0.07834794372320175,
|
210 |
+
"learning_rate": 0.0001658401584698049,
|
211 |
+
"loss": 0.8419,
|
212 |
+
"step": 140
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.35714285714285715,
|
216 |
+
"grad_norm": 0.08485198020935059,
|
217 |
+
"learning_rate": 0.00016254105729852464,
|
218 |
+
"loss": 0.8735,
|
219 |
+
"step": 145
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.3694581280788177,
|
223 |
+
"grad_norm": 0.07340410351753235,
|
224 |
+
"learning_rate": 0.00015912614448635782,
|
225 |
+
"loss": 0.843,
|
226 |
+
"step": 150
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.3817733990147783,
|
230 |
+
"grad_norm": 0.07572367042303085,
|
231 |
+
"learning_rate": 0.00015560174366570446,
|
232 |
+
"loss": 0.8396,
|
233 |
+
"step": 155
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.39408866995073893,
|
237 |
+
"grad_norm": 0.07385943084955215,
|
238 |
+
"learning_rate": 0.0001519743812155516,
|
239 |
+
"loss": 0.8658,
|
240 |
+
"step": 160
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.4064039408866995,
|
244 |
+
"grad_norm": 0.07291774451732635,
|
245 |
+
"learning_rate": 0.00014825077417612186,
|
246 |
+
"loss": 0.8571,
|
247 |
+
"step": 165
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.4187192118226601,
|
251 |
+
"grad_norm": 0.0746372640132904,
|
252 |
+
"learning_rate": 0.00014443781781046136,
|
253 |
+
"loss": 0.8532,
|
254 |
+
"step": 170
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.43103448275862066,
|
258 |
+
"grad_norm": 0.07541359215974808,
|
259 |
+
"learning_rate": 0.00014054257283599973,
|
260 |
+
"loss": 0.8461,
|
261 |
+
"step": 175
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.4433497536945813,
|
265 |
+
"grad_norm": 0.0773872435092926,
|
266 |
+
"learning_rate": 0.00013657225234972695,
|
267 |
+
"loss": 0.8506,
|
268 |
+
"step": 180
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.45566502463054187,
|
272 |
+
"grad_norm": 0.07032646983861923,
|
273 |
+
"learning_rate": 0.00013253420847119803,
|
274 |
+
"loss": 0.859,
|
275 |
+
"step": 185
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.46798029556650245,
|
279 |
+
"grad_norm": 0.07974265515804291,
|
280 |
+
"learning_rate": 0.0001284359187281004,
|
281 |
+
"loss": 0.8493,
|
282 |
+
"step": 190
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.4802955665024631,
|
286 |
+
"grad_norm": 0.09293660521507263,
|
287 |
+
"learning_rate": 0.0001242849722095936,
|
288 |
+
"loss": 0.8355,
|
289 |
+
"step": 195
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.49261083743842365,
|
293 |
+
"grad_norm": 0.08410409837961197,
|
294 |
+
"learning_rate": 0.00012008905551306356,
|
295 |
+
"loss": 0.8214,
|
296 |
+
"step": 200
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.5049261083743842,
|
300 |
+
"grad_norm": 0.08014179021120071,
|
301 |
+
"learning_rate": 0.00011585593851031347,
|
302 |
+
"loss": 0.8358,
|
303 |
+
"step": 205
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.5172413793103449,
|
307 |
+
"grad_norm": 0.09226740896701813,
|
308 |
+
"learning_rate": 0.00011159345995955006,
|
309 |
+
"loss": 0.8496,
|
310 |
+
"step": 210
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.5295566502463054,
|
314 |
+
"grad_norm": 0.08234479278326035,
|
315 |
+
"learning_rate": 0.00010730951298980776,
|
316 |
+
"loss": 0.8604,
|
317 |
+
"step": 215
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.541871921182266,
|
321 |
+
"grad_norm": 0.08139283210039139,
|
322 |
+
"learning_rate": 0.00010301203048469083,
|
323 |
+
"loss": 0.8538,
|
324 |
+
"step": 220
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.5541871921182266,
|
328 |
+
"grad_norm": 0.0822574719786644,
|
329 |
+
"learning_rate": 9.870897039249911e-05,
|
330 |
+
"loss": 0.8394,
|
331 |
+
"step": 225
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.5665024630541872,
|
335 |
+
"grad_norm": 0.07834915816783905,
|
336 |
+
"learning_rate": 9.440830098993969e-05,
|
337 |
+
"loss": 0.8513,
|
338 |
+
"step": 230
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.5788177339901478,
|
342 |
+
"grad_norm": 0.08242921531200409,
|
343 |
+
"learning_rate": 9.011798612671286e-05,
|
344 |
+
"loss": 0.846,
|
345 |
+
"step": 235
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.5911330049261084,
|
349 |
+
"grad_norm": 0.07460321485996246,
|
350 |
+
"learning_rate": 8.58459704782957e-05,
|
351 |
+
"loss": 0.856,
|
352 |
+
"step": 240
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.603448275862069,
|
356 |
+
"grad_norm": 0.08328606188297272,
|
357 |
+
"learning_rate": 8.160016483423199e-05,
|
358 |
+
"loss": 0.8537,
|
359 |
+
"step": 245
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.6157635467980296,
|
363 |
+
"grad_norm": 0.0797269344329834,
|
364 |
+
"learning_rate": 7.738843144917119e-05,
|
365 |
+
"loss": 0.8439,
|
366 |
+
"step": 250
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.6280788177339901,
|
370 |
+
"grad_norm": 0.07935849577188492,
|
371 |
+
"learning_rate": 7.321856948378259e-05,
|
372 |
+
"loss": 0.8358,
|
373 |
+
"step": 255
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.6403940886699507,
|
377 |
+
"grad_norm": 0.07458296418190002,
|
378 |
+
"learning_rate": 6.909830056250527e-05,
|
379 |
+
"loss": 0.8485,
|
380 |
+
"step": 260
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.6527093596059114,
|
384 |
+
"grad_norm": 0.07776927202939987,
|
385 |
+
"learning_rate": 6.503525447487715e-05,
|
386 |
+
"loss": 0.8273,
|
387 |
+
"step": 265
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.6650246305418719,
|
391 |
+
"grad_norm": 0.08262035250663757,
|
392 |
+
"learning_rate": 6.103695504692122e-05,
|
393 |
+
"loss": 0.8261,
|
394 |
+
"step": 270
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.6773399014778325,
|
398 |
+
"grad_norm": 0.0770513042807579,
|
399 |
+
"learning_rate": 5.7110806208751655e-05,
|
400 |
+
"loss": 0.8404,
|
401 |
+
"step": 275
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.6896551724137931,
|
405 |
+
"grad_norm": 0.08033966273069382,
|
406 |
+
"learning_rate": 5.326407828419979e-05,
|
407 |
+
"loss": 0.832,
|
408 |
+
"step": 280
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.7019704433497537,
|
412 |
+
"grad_norm": 0.08418203890323639,
|
413 |
+
"learning_rate": 4.9503894527847964e-05,
|
414 |
+
"loss": 0.8397,
|
415 |
+
"step": 285
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.7142857142857143,
|
419 |
+
"grad_norm": 0.07704119384288788,
|
420 |
+
"learning_rate": 4.583721793440188e-05,
|
421 |
+
"loss": 0.8865,
|
422 |
+
"step": 290
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.7266009852216748,
|
426 |
+
"grad_norm": 0.07464379817247391,
|
427 |
+
"learning_rate": 4.227083834482728e-05,
|
428 |
+
"loss": 0.8402,
|
429 |
+
"step": 295
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.7389162561576355,
|
433 |
+
"grad_norm": 0.08190499991178513,
|
434 |
+
"learning_rate": 3.881135987312757e-05,
|
435 |
+
"loss": 0.8428,
|
436 |
+
"step": 300
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.7512315270935961,
|
440 |
+
"grad_norm": 0.07950634509325027,
|
441 |
+
"learning_rate": 3.546518867704499e-05,
|
442 |
+
"loss": 0.8559,
|
443 |
+
"step": 305
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.7635467980295566,
|
447 |
+
"grad_norm": 0.08059704303741455,
|
448 |
+
"learning_rate": 3.223852109533112e-05,
|
449 |
+
"loss": 0.8514,
|
450 |
+
"step": 310
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.7758620689655172,
|
454 |
+
"grad_norm": 0.07899323105812073,
|
455 |
+
"learning_rate": 2.9137332173554043e-05,
|
456 |
+
"loss": 0.8284,
|
457 |
+
"step": 315
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.7881773399014779,
|
461 |
+
"grad_norm": 0.08275061845779419,
|
462 |
+
"learning_rate": 2.616736459968936e-05,
|
463 |
+
"loss": 0.8311,
|
464 |
+
"step": 320
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.8004926108374384,
|
468 |
+
"grad_norm": 0.07869672030210495,
|
469 |
+
"learning_rate": 2.33341180699841e-05,
|
470 |
+
"loss": 0.8332,
|
471 |
+
"step": 325
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.812807881773399,
|
475 |
+
"grad_norm": 0.07721079885959625,
|
476 |
+
"learning_rate": 2.0642839104785272e-05,
|
477 |
+
"loss": 0.8342,
|
478 |
+
"step": 330
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.8251231527093597,
|
482 |
+
"grad_norm": 0.08025462925434113,
|
483 |
+
"learning_rate": 1.8098511333192024e-05,
|
484 |
+
"loss": 0.8573,
|
485 |
+
"step": 335
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.8374384236453202,
|
489 |
+
"grad_norm": 0.07781907171010971,
|
490 |
+
"learning_rate": 1.570584626452173e-05,
|
491 |
+
"loss": 0.8356,
|
492 |
+
"step": 340
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.8497536945812808,
|
496 |
+
"grad_norm": 0.07266338169574738,
|
497 |
+
"learning_rate": 1.3469274563679402e-05,
|
498 |
+
"loss": 0.8222,
|
499 |
+
"step": 345
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.8620689655172413,
|
503 |
+
"grad_norm": 0.07625401765108109,
|
504 |
+
"learning_rate": 1.1392937846586215e-05,
|
505 |
+
"loss": 0.8371,
|
506 |
+
"step": 350
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.874384236453202,
|
510 |
+
"grad_norm": 0.07775253057479858,
|
511 |
+
"learning_rate": 9.48068101086026e-06,
|
512 |
+
"loss": 0.8349,
|
513 |
+
"step": 355
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.8866995073891626,
|
517 |
+
"grad_norm": 0.07747705280780792,
|
518 |
+
"learning_rate": 7.736045115951251e-06,
|
519 |
+
"loss": 0.8512,
|
520 |
+
"step": 360
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.8990147783251231,
|
524 |
+
"grad_norm": 0.08055680245161057,
|
525 |
+
"learning_rate": 6.16226082591359e-06,
|
526 |
+
"loss": 0.8345,
|
527 |
+
"step": 365
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.9113300492610837,
|
531 |
+
"grad_norm": 0.07874856889247894,
|
532 |
+
"learning_rate": 4.762242426960262e-06,
|
533 |
+
"loss": 0.8484,
|
534 |
+
"step": 370
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.9236453201970444,
|
538 |
+
"grad_norm": 0.08116665482521057,
|
539 |
+
"learning_rate": 3.5385824308756587e-06,
|
540 |
+
"loss": 0.8194,
|
541 |
+
"step": 375
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.9359605911330049,
|
545 |
+
"grad_norm": 0.08248580247163773,
|
546 |
+
"learning_rate": 2.493546774280531e-06,
|
547 |
+
"loss": 0.8563,
|
548 |
+
"step": 380
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.9482758620689655,
|
552 |
+
"grad_norm": 0.08016284555196762,
|
553 |
+
"learning_rate": 1.6290706226390285e-06,
|
554 |
+
"loss": 0.8388,
|
555 |
+
"step": 385
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.9605911330049262,
|
559 |
+
"grad_norm": 0.07426544278860092,
|
560 |
+
"learning_rate": 9.46754786777726e-07,
|
561 |
+
"loss": 0.8186,
|
562 |
+
"step": 390
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.9729064039408867,
|
566 |
+
"grad_norm": 0.07471868395805359,
|
567 |
+
"learning_rate": 4.4786275855247527e-07,
|
568 |
+
"loss": 0.8358,
|
569 |
+
"step": 395
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.9852216748768473,
|
573 |
+
"grad_norm": 0.0780901238322258,
|
574 |
+
"learning_rate": 1.333183711524133e-07,
|
575 |
+
"loss": 0.8316,
|
576 |
+
"step": 400
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.9975369458128078,
|
580 |
+
"grad_norm": 0.07565240561962128,
|
581 |
+
"learning_rate": 3.7040883734462683e-09,
|
582 |
+
"loss": 0.8215,
|
583 |
+
"step": 405
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 1.0,
|
587 |
+
"eval_loss": 1.0929372310638428,
|
588 |
+
"eval_runtime": 133.5096,
|
589 |
+
"eval_samples_per_second": 8.651,
|
590 |
+
"eval_steps_per_second": 0.547,
|
591 |
+
"step": 406
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 1.0,
|
595 |
+
"step": 406,
|
596 |
+
"total_flos": 1.9785163757689242e+18,
|
597 |
+
"train_loss": 0.8540300870763844,
|
598 |
+
"train_runtime": 31542.2056,
|
599 |
+
"train_samples_per_second": 3.295,
|
600 |
+
"train_steps_per_second": 0.013
|
601 |
+
}
|
602 |
+
],
|
603 |
+
"logging_steps": 5,
|
604 |
+
"max_steps": 406,
|
605 |
+
"num_input_tokens_seen": 0,
|
606 |
+
"num_train_epochs": 1,
|
607 |
+
"save_steps": 25,
|
608 |
+
"stateful_callbacks": {
|
609 |
+
"TrainerControl": {
|
610 |
+
"args": {
|
611 |
+
"should_epoch_stop": false,
|
612 |
+
"should_evaluate": false,
|
613 |
+
"should_log": false,
|
614 |
+
"should_save": true,
|
615 |
+
"should_training_stop": false
|
616 |
+
},
|
617 |
+
"attributes": {}
|
618 |
+
}
|
619 |
+
},
|
620 |
+
"total_flos": 1.9785163757689242e+18,
|
621 |
+
"train_batch_size": 8,
|
622 |
+
"trial_name": null,
|
623 |
+
"trial_params": null
|
624 |
+
}
|