End of training
Browse files- README.md +2 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +640 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
|
|
4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
5 |
tags:
|
6 |
- llama-factory
|
|
|
7 |
- generated_from_trainer
|
8 |
model-index:
|
9 |
- name: oh_v1.3_opengpt_x8
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
# oh_v1.3_opengpt_x8
|
17 |
|
18 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on
|
19 |
It achieves the following results on the evaluation set:
|
20 |
- Loss: 0.7300
|
21 |
|
|
|
4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
5 |
tags:
|
6 |
- llama-factory
|
7 |
+
- full
|
8 |
- generated_from_trainer
|
9 |
model-index:
|
10 |
- name: oh_v1.3_opengpt_x8
|
|
|
16 |
|
17 |
# oh_v1.3_opengpt_x8
|
18 |
|
19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/oh_v1.3_opengpt_x8 dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
- Loss: 0.7300
|
22 |
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_loss": 0.7299705147743225,
|
4 |
+
"eval_runtime": 26.5424,
|
5 |
+
"eval_samples_per_second": 279.101,
|
6 |
+
"eval_steps_per_second": 1.093,
|
7 |
+
"total_flos": 1381905727488000.0,
|
8 |
+
"train_loss": 0.7044297796307188,
|
9 |
+
"train_runtime": 5353.5806,
|
10 |
+
"train_samples_per_second": 78.866,
|
11 |
+
"train_steps_per_second": 0.154
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_loss": 0.7299705147743225,
|
4 |
+
"eval_runtime": 26.5424,
|
5 |
+
"eval_samples_per_second": 279.101,
|
6 |
+
"eval_steps_per_second": 1.093
|
7 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"total_flos": 1381905727488000.0,
|
4 |
+
"train_loss": 0.7044297796307188,
|
5 |
+
"train_runtime": 5353.5806,
|
6 |
+
"train_samples_per_second": 78.866,
|
7 |
+
"train_steps_per_second": 0.154
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,640 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 3.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 825,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.03636363636363636,
|
13 |
+
"grad_norm": 3.2293971017711174,
|
14 |
+
"learning_rate": 5e-06,
|
15 |
+
"loss": 1.0337,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.07272727272727272,
|
20 |
+
"grad_norm": 1.2461654883314972,
|
21 |
+
"learning_rate": 5e-06,
|
22 |
+
"loss": 0.9092,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.10909090909090909,
|
27 |
+
"grad_norm": 1.0937923635217501,
|
28 |
+
"learning_rate": 5e-06,
|
29 |
+
"loss": 0.8658,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.14545454545454545,
|
34 |
+
"grad_norm": 1.3350225945199414,
|
35 |
+
"learning_rate": 5e-06,
|
36 |
+
"loss": 0.844,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.18181818181818182,
|
41 |
+
"grad_norm": 1.029425810987488,
|
42 |
+
"learning_rate": 5e-06,
|
43 |
+
"loss": 0.8249,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.21818181818181817,
|
48 |
+
"grad_norm": 1.5219290967515304,
|
49 |
+
"learning_rate": 5e-06,
|
50 |
+
"loss": 0.8068,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.2545454545454545,
|
55 |
+
"grad_norm": 1.6740495880819521,
|
56 |
+
"learning_rate": 5e-06,
|
57 |
+
"loss": 0.7989,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.2909090909090909,
|
62 |
+
"grad_norm": 1.2973735477904815,
|
63 |
+
"learning_rate": 5e-06,
|
64 |
+
"loss": 0.7921,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.32727272727272727,
|
69 |
+
"grad_norm": 0.8566363002967183,
|
70 |
+
"learning_rate": 5e-06,
|
71 |
+
"loss": 0.781,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.36363636363636365,
|
76 |
+
"grad_norm": 0.9961672641644985,
|
77 |
+
"learning_rate": 5e-06,
|
78 |
+
"loss": 0.7745,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.4,
|
83 |
+
"grad_norm": 0.8927257959121373,
|
84 |
+
"learning_rate": 5e-06,
|
85 |
+
"loss": 0.7754,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.43636363636363634,
|
90 |
+
"grad_norm": 0.6135178704985191,
|
91 |
+
"learning_rate": 5e-06,
|
92 |
+
"loss": 0.772,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.4727272727272727,
|
97 |
+
"grad_norm": 0.7431505188106242,
|
98 |
+
"learning_rate": 5e-06,
|
99 |
+
"loss": 0.7686,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.509090909090909,
|
104 |
+
"grad_norm": 0.7150787812569424,
|
105 |
+
"learning_rate": 5e-06,
|
106 |
+
"loss": 0.7618,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.5454545454545454,
|
111 |
+
"grad_norm": 0.6352342662453642,
|
112 |
+
"learning_rate": 5e-06,
|
113 |
+
"loss": 0.7611,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.5818181818181818,
|
118 |
+
"grad_norm": 0.6257901300873526,
|
119 |
+
"learning_rate": 5e-06,
|
120 |
+
"loss": 0.7569,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.6181818181818182,
|
125 |
+
"grad_norm": 0.6387102446786417,
|
126 |
+
"learning_rate": 5e-06,
|
127 |
+
"loss": 0.7611,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.6545454545454545,
|
132 |
+
"grad_norm": 0.5983754152683597,
|
133 |
+
"learning_rate": 5e-06,
|
134 |
+
"loss": 0.7546,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.6909090909090909,
|
139 |
+
"grad_norm": 0.7480127979666656,
|
140 |
+
"learning_rate": 5e-06,
|
141 |
+
"loss": 0.7566,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.7272727272727273,
|
146 |
+
"grad_norm": 0.5804396007389026,
|
147 |
+
"learning_rate": 5e-06,
|
148 |
+
"loss": 0.75,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.7636363636363637,
|
153 |
+
"grad_norm": 0.682148918886327,
|
154 |
+
"learning_rate": 5e-06,
|
155 |
+
"loss": 0.7476,
|
156 |
+
"step": 210
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.8,
|
160 |
+
"grad_norm": 0.8039336411015884,
|
161 |
+
"learning_rate": 5e-06,
|
162 |
+
"loss": 0.7462,
|
163 |
+
"step": 220
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.8363636363636363,
|
167 |
+
"grad_norm": 0.6876607052536684,
|
168 |
+
"learning_rate": 5e-06,
|
169 |
+
"loss": 0.7411,
|
170 |
+
"step": 230
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.8727272727272727,
|
174 |
+
"grad_norm": 0.6588151842699974,
|
175 |
+
"learning_rate": 5e-06,
|
176 |
+
"loss": 0.7469,
|
177 |
+
"step": 240
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.9090909090909091,
|
181 |
+
"grad_norm": 0.6715213794720472,
|
182 |
+
"learning_rate": 5e-06,
|
183 |
+
"loss": 0.7378,
|
184 |
+
"step": 250
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.9454545454545454,
|
188 |
+
"grad_norm": 0.5870957383826958,
|
189 |
+
"learning_rate": 5e-06,
|
190 |
+
"loss": 0.7457,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.9818181818181818,
|
195 |
+
"grad_norm": 0.6643986810785624,
|
196 |
+
"learning_rate": 5e-06,
|
197 |
+
"loss": 0.7466,
|
198 |
+
"step": 270
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 1.0,
|
202 |
+
"eval_loss": 0.7417545914649963,
|
203 |
+
"eval_runtime": 26.6218,
|
204 |
+
"eval_samples_per_second": 278.268,
|
205 |
+
"eval_steps_per_second": 1.089,
|
206 |
+
"step": 275
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 1.018181818181818,
|
210 |
+
"grad_norm": 0.9781239153342394,
|
211 |
+
"learning_rate": 5e-06,
|
212 |
+
"loss": 0.7136,
|
213 |
+
"step": 280
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 1.0545454545454545,
|
217 |
+
"grad_norm": 0.7152925984087143,
|
218 |
+
"learning_rate": 5e-06,
|
219 |
+
"loss": 0.6871,
|
220 |
+
"step": 290
|
221 |
+
},
|
222 |
+
{
|
223 |
+
"epoch": 1.0909090909090908,
|
224 |
+
"grad_norm": 0.6929492576277494,
|
225 |
+
"learning_rate": 5e-06,
|
226 |
+
"loss": 0.6894,
|
227 |
+
"step": 300
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 1.1272727272727272,
|
231 |
+
"grad_norm": 0.728764264622129,
|
232 |
+
"learning_rate": 5e-06,
|
233 |
+
"loss": 0.6935,
|
234 |
+
"step": 310
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 1.1636363636363636,
|
238 |
+
"grad_norm": 0.7252517543389313,
|
239 |
+
"learning_rate": 5e-06,
|
240 |
+
"loss": 0.6945,
|
241 |
+
"step": 320
|
242 |
+
},
|
243 |
+
{
|
244 |
+
"epoch": 1.2,
|
245 |
+
"grad_norm": 0.6665160391388197,
|
246 |
+
"learning_rate": 5e-06,
|
247 |
+
"loss": 0.6886,
|
248 |
+
"step": 330
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 1.2363636363636363,
|
252 |
+
"grad_norm": 0.7161659905517039,
|
253 |
+
"learning_rate": 5e-06,
|
254 |
+
"loss": 0.6898,
|
255 |
+
"step": 340
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 1.2727272727272727,
|
259 |
+
"grad_norm": 0.5719039452566653,
|
260 |
+
"learning_rate": 5e-06,
|
261 |
+
"loss": 0.6934,
|
262 |
+
"step": 350
|
263 |
+
},
|
264 |
+
{
|
265 |
+
"epoch": 1.309090909090909,
|
266 |
+
"grad_norm": 0.6060853746189843,
|
267 |
+
"learning_rate": 5e-06,
|
268 |
+
"loss": 0.6922,
|
269 |
+
"step": 360
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 1.3454545454545455,
|
273 |
+
"grad_norm": 0.6563719933283224,
|
274 |
+
"learning_rate": 5e-06,
|
275 |
+
"loss": 0.6912,
|
276 |
+
"step": 370
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 1.3818181818181818,
|
280 |
+
"grad_norm": 0.5958006047997326,
|
281 |
+
"learning_rate": 5e-06,
|
282 |
+
"loss": 0.6904,
|
283 |
+
"step": 380
|
284 |
+
},
|
285 |
+
{
|
286 |
+
"epoch": 1.4181818181818182,
|
287 |
+
"grad_norm": 0.7430218105320606,
|
288 |
+
"learning_rate": 5e-06,
|
289 |
+
"loss": 0.688,
|
290 |
+
"step": 390
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 1.4545454545454546,
|
294 |
+
"grad_norm": 0.6322073230662588,
|
295 |
+
"learning_rate": 5e-06,
|
296 |
+
"loss": 0.6883,
|
297 |
+
"step": 400
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 1.490909090909091,
|
301 |
+
"grad_norm": 0.7151221978666452,
|
302 |
+
"learning_rate": 5e-06,
|
303 |
+
"loss": 0.6934,
|
304 |
+
"step": 410
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 1.5272727272727273,
|
308 |
+
"grad_norm": 0.6184168187218901,
|
309 |
+
"learning_rate": 5e-06,
|
310 |
+
"loss": 0.6916,
|
311 |
+
"step": 420
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 1.5636363636363635,
|
315 |
+
"grad_norm": 0.6280848540221795,
|
316 |
+
"learning_rate": 5e-06,
|
317 |
+
"loss": 0.6916,
|
318 |
+
"step": 430
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 1.6,
|
322 |
+
"grad_norm": 0.6568705155050817,
|
323 |
+
"learning_rate": 5e-06,
|
324 |
+
"loss": 0.6856,
|
325 |
+
"step": 440
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 1.6363636363636362,
|
329 |
+
"grad_norm": 0.6359258851827682,
|
330 |
+
"learning_rate": 5e-06,
|
331 |
+
"loss": 0.6851,
|
332 |
+
"step": 450
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 1.6727272727272728,
|
336 |
+
"grad_norm": 0.710888538426671,
|
337 |
+
"learning_rate": 5e-06,
|
338 |
+
"loss": 0.6872,
|
339 |
+
"step": 460
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 1.709090909090909,
|
343 |
+
"grad_norm": 0.7584066029266229,
|
344 |
+
"learning_rate": 5e-06,
|
345 |
+
"loss": 0.6849,
|
346 |
+
"step": 470
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 1.7454545454545456,
|
350 |
+
"grad_norm": 0.5960492892442344,
|
351 |
+
"learning_rate": 5e-06,
|
352 |
+
"loss": 0.6891,
|
353 |
+
"step": 480
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 1.7818181818181817,
|
357 |
+
"grad_norm": 0.5629377755020811,
|
358 |
+
"learning_rate": 5e-06,
|
359 |
+
"loss": 0.6847,
|
360 |
+
"step": 490
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 1.8181818181818183,
|
364 |
+
"grad_norm": 0.589716689792314,
|
365 |
+
"learning_rate": 5e-06,
|
366 |
+
"loss": 0.6871,
|
367 |
+
"step": 500
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 1.8545454545454545,
|
371 |
+
"grad_norm": 0.5740509121739076,
|
372 |
+
"learning_rate": 5e-06,
|
373 |
+
"loss": 0.6888,
|
374 |
+
"step": 510
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 1.8909090909090909,
|
378 |
+
"grad_norm": 0.5891046247600111,
|
379 |
+
"learning_rate": 5e-06,
|
380 |
+
"loss": 0.6884,
|
381 |
+
"step": 520
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 1.9272727272727272,
|
385 |
+
"grad_norm": 0.6447276827053491,
|
386 |
+
"learning_rate": 5e-06,
|
387 |
+
"loss": 0.6893,
|
388 |
+
"step": 530
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 1.9636363636363636,
|
392 |
+
"grad_norm": 0.6935516132206995,
|
393 |
+
"learning_rate": 5e-06,
|
394 |
+
"loss": 0.6868,
|
395 |
+
"step": 540
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 2.0,
|
399 |
+
"grad_norm": 0.5781509823001448,
|
400 |
+
"learning_rate": 5e-06,
|
401 |
+
"loss": 0.6841,
|
402 |
+
"step": 550
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 2.0,
|
406 |
+
"eval_loss": 0.7281343340873718,
|
407 |
+
"eval_runtime": 26.4698,
|
408 |
+
"eval_samples_per_second": 279.867,
|
409 |
+
"eval_steps_per_second": 1.096,
|
410 |
+
"step": 550
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 2.036363636363636,
|
414 |
+
"grad_norm": 0.7551729949207574,
|
415 |
+
"learning_rate": 5e-06,
|
416 |
+
"loss": 0.6351,
|
417 |
+
"step": 560
|
418 |
+
},
|
419 |
+
{
|
420 |
+
"epoch": 2.0727272727272728,
|
421 |
+
"grad_norm": 0.6070448901420726,
|
422 |
+
"learning_rate": 5e-06,
|
423 |
+
"loss": 0.6307,
|
424 |
+
"step": 570
|
425 |
+
},
|
426 |
+
{
|
427 |
+
"epoch": 2.109090909090909,
|
428 |
+
"grad_norm": 0.7225948313371118,
|
429 |
+
"learning_rate": 5e-06,
|
430 |
+
"loss": 0.6357,
|
431 |
+
"step": 580
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 2.1454545454545455,
|
435 |
+
"grad_norm": 1.3944109200671733,
|
436 |
+
"learning_rate": 5e-06,
|
437 |
+
"loss": 0.6375,
|
438 |
+
"step": 590
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 2.1818181818181817,
|
442 |
+
"grad_norm": 1.1390572133302885,
|
443 |
+
"learning_rate": 5e-06,
|
444 |
+
"loss": 0.635,
|
445 |
+
"step": 600
|
446 |
+
},
|
447 |
+
{
|
448 |
+
"epoch": 2.2181818181818183,
|
449 |
+
"grad_norm": 0.7900509422330505,
|
450 |
+
"learning_rate": 5e-06,
|
451 |
+
"loss": 0.6383,
|
452 |
+
"step": 610
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 2.2545454545454544,
|
456 |
+
"grad_norm": 0.594871030626621,
|
457 |
+
"learning_rate": 5e-06,
|
458 |
+
"loss": 0.6321,
|
459 |
+
"step": 620
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 2.290909090909091,
|
463 |
+
"grad_norm": 0.665898906007086,
|
464 |
+
"learning_rate": 5e-06,
|
465 |
+
"loss": 0.6341,
|
466 |
+
"step": 630
|
467 |
+
},
|
468 |
+
{
|
469 |
+
"epoch": 2.327272727272727,
|
470 |
+
"grad_norm": 0.6509722897169726,
|
471 |
+
"learning_rate": 5e-06,
|
472 |
+
"loss": 0.6326,
|
473 |
+
"step": 640
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 2.3636363636363638,
|
477 |
+
"grad_norm": 0.6231670817005929,
|
478 |
+
"learning_rate": 5e-06,
|
479 |
+
"loss": 0.6385,
|
480 |
+
"step": 650
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 2.4,
|
484 |
+
"grad_norm": 0.6425410588561774,
|
485 |
+
"learning_rate": 5e-06,
|
486 |
+
"loss": 0.6373,
|
487 |
+
"step": 660
|
488 |
+
},
|
489 |
+
{
|
490 |
+
"epoch": 2.4363636363636365,
|
491 |
+
"grad_norm": 0.621241338432262,
|
492 |
+
"learning_rate": 5e-06,
|
493 |
+
"loss": 0.6399,
|
494 |
+
"step": 670
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 2.4727272727272727,
|
498 |
+
"grad_norm": 0.6924233110335524,
|
499 |
+
"learning_rate": 5e-06,
|
500 |
+
"loss": 0.6393,
|
501 |
+
"step": 680
|
502 |
+
},
|
503 |
+
{
|
504 |
+
"epoch": 2.509090909090909,
|
505 |
+
"grad_norm": 0.6419114963815122,
|
506 |
+
"learning_rate": 5e-06,
|
507 |
+
"loss": 0.6405,
|
508 |
+
"step": 690
|
509 |
+
},
|
510 |
+
{
|
511 |
+
"epoch": 2.5454545454545454,
|
512 |
+
"grad_norm": 0.7336852368102121,
|
513 |
+
"learning_rate": 5e-06,
|
514 |
+
"loss": 0.6385,
|
515 |
+
"step": 700
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 2.581818181818182,
|
519 |
+
"grad_norm": 0.7922288944252411,
|
520 |
+
"learning_rate": 5e-06,
|
521 |
+
"loss": 0.6377,
|
522 |
+
"step": 710
|
523 |
+
},
|
524 |
+
{
|
525 |
+
"epoch": 2.618181818181818,
|
526 |
+
"grad_norm": 0.6500377491351792,
|
527 |
+
"learning_rate": 5e-06,
|
528 |
+
"loss": 0.6427,
|
529 |
+
"step": 720
|
530 |
+
},
|
531 |
+
{
|
532 |
+
"epoch": 2.6545454545454543,
|
533 |
+
"grad_norm": 0.6853834065254241,
|
534 |
+
"learning_rate": 5e-06,
|
535 |
+
"loss": 0.6346,
|
536 |
+
"step": 730
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 2.690909090909091,
|
540 |
+
"grad_norm": 0.8156333668312422,
|
541 |
+
"learning_rate": 5e-06,
|
542 |
+
"loss": 0.6414,
|
543 |
+
"step": 740
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 2.7272727272727275,
|
547 |
+
"grad_norm": 0.6294215183471213,
|
548 |
+
"learning_rate": 5e-06,
|
549 |
+
"loss": 0.6363,
|
550 |
+
"step": 750
|
551 |
+
},
|
552 |
+
{
|
553 |
+
"epoch": 2.7636363636363637,
|
554 |
+
"grad_norm": 0.8237171162592375,
|
555 |
+
"learning_rate": 5e-06,
|
556 |
+
"loss": 0.6421,
|
557 |
+
"step": 760
|
558 |
+
},
|
559 |
+
{
|
560 |
+
"epoch": 2.8,
|
561 |
+
"grad_norm": 0.6772752476166749,
|
562 |
+
"learning_rate": 5e-06,
|
563 |
+
"loss": 0.6356,
|
564 |
+
"step": 770
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 2.8363636363636364,
|
568 |
+
"grad_norm": 0.7780500988065099,
|
569 |
+
"learning_rate": 5e-06,
|
570 |
+
"loss": 0.6425,
|
571 |
+
"step": 780
|
572 |
+
},
|
573 |
+
{
|
574 |
+
"epoch": 2.8727272727272726,
|
575 |
+
"grad_norm": 0.6862874007983163,
|
576 |
+
"learning_rate": 5e-06,
|
577 |
+
"loss": 0.6368,
|
578 |
+
"step": 790
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 2.909090909090909,
|
582 |
+
"grad_norm": 0.5748210856771035,
|
583 |
+
"learning_rate": 5e-06,
|
584 |
+
"loss": 0.6405,
|
585 |
+
"step": 800
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 2.9454545454545453,
|
589 |
+
"grad_norm": 0.6351457621560951,
|
590 |
+
"learning_rate": 5e-06,
|
591 |
+
"loss": 0.6357,
|
592 |
+
"step": 810
|
593 |
+
},
|
594 |
+
{
|
595 |
+
"epoch": 2.981818181818182,
|
596 |
+
"grad_norm": 0.586627253325874,
|
597 |
+
"learning_rate": 5e-06,
|
598 |
+
"loss": 0.6412,
|
599 |
+
"step": 820
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 3.0,
|
603 |
+
"eval_loss": 0.7299705147743225,
|
604 |
+
"eval_runtime": 25.9228,
|
605 |
+
"eval_samples_per_second": 285.772,
|
606 |
+
"eval_steps_per_second": 1.119,
|
607 |
+
"step": 825
|
608 |
+
},
|
609 |
+
{
|
610 |
+
"epoch": 3.0,
|
611 |
+
"step": 825,
|
612 |
+
"total_flos": 1381905727488000.0,
|
613 |
+
"train_loss": 0.7044297796307188,
|
614 |
+
"train_runtime": 5353.5806,
|
615 |
+
"train_samples_per_second": 78.866,
|
616 |
+
"train_steps_per_second": 0.154
|
617 |
+
}
|
618 |
+
],
|
619 |
+
"logging_steps": 10,
|
620 |
+
"max_steps": 825,
|
621 |
+
"num_input_tokens_seen": 0,
|
622 |
+
"num_train_epochs": 3,
|
623 |
+
"save_steps": 500,
|
624 |
+
"stateful_callbacks": {
|
625 |
+
"TrainerControl": {
|
626 |
+
"args": {
|
627 |
+
"should_epoch_stop": false,
|
628 |
+
"should_evaluate": false,
|
629 |
+
"should_log": false,
|
630 |
+
"should_save": true,
|
631 |
+
"should_training_stop": true
|
632 |
+
},
|
633 |
+
"attributes": {}
|
634 |
+
}
|
635 |
+
},
|
636 |
+
"total_flos": 1381905727488000.0,
|
637 |
+
"train_batch_size": 16,
|
638 |
+
"trial_name": null,
|
639 |
+
"trial_params": null
|
640 |
+
}
|
training_eval_loss.png
ADDED
training_loss.png
ADDED