MHGanainy/gpt2-xl-lora-ecthr-random-imbalanced-skewed-cluster-8-id-6
Browse files- README.md +2 -0
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +644 -0
README.md
CHANGED
@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
# gpt2-xl-lora-ecthr-random-imbalanced-skewed-cluster-8-id-6
|
16 |
|
17 |
This model is a fine-tuned version of [openai-community/gpt2-xl](https://huggingface.co/openai-community/gpt2-xl) on an unknown dataset.
|
|
|
|
|
18 |
|
19 |
## Model description
|
20 |
|
|
|
15 |
# gpt2-xl-lora-ecthr-random-imbalanced-skewed-cluster-8-id-6
|
16 |
|
17 |
This model is a fine-tuned version of [openai-community/gpt2-xl](https://huggingface.co/openai-community/gpt2-xl) on an unknown dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 1.9178
|
20 |
|
21 |
## Model description
|
22 |
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_loss": 1.9178457260131836,
|
4 |
+
"eval_runtime": 80.5533,
|
5 |
+
"eval_samples_per_second": 21.774,
|
6 |
+
"eval_steps_per_second": 2.731,
|
7 |
+
"perplexity": 6.806280072216437,
|
8 |
+
"total_flos": 1.58472345255936e+17,
|
9 |
+
"train_loss": 2.0172890398896968,
|
10 |
+
"train_runtime": 2382.4555,
|
11 |
+
"train_samples_per_second": 7.283,
|
12 |
+
"train_steps_per_second": 3.642
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_loss": 1.9178457260131836,
|
4 |
+
"eval_runtime": 80.5533,
|
5 |
+
"eval_samples_per_second": 21.774,
|
6 |
+
"eval_steps_per_second": 2.731,
|
7 |
+
"perplexity": 6.806280072216437
|
8 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"total_flos": 1.58472345255936e+17,
|
4 |
+
"train_loss": 2.0172890398896968,
|
5 |
+
"train_runtime": 2382.4555,
|
6 |
+
"train_samples_per_second": 7.283,
|
7 |
+
"train_steps_per_second": 3.642
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,644 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 8676,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.011526048870447211,
|
13 |
+
"grad_norm": 0.19771268963813782,
|
14 |
+
"learning_rate": 2.0000000000000003e-06,
|
15 |
+
"loss": 2.4498,
|
16 |
+
"step": 100
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.023052097740894423,
|
20 |
+
"grad_norm": 0.16720762848854065,
|
21 |
+
"learning_rate": 4.000000000000001e-06,
|
22 |
+
"loss": 2.4238,
|
23 |
+
"step": 200
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.034578146611341634,
|
27 |
+
"grad_norm": 0.2238711267709732,
|
28 |
+
"learning_rate": 6e-06,
|
29 |
+
"loss": 2.3975,
|
30 |
+
"step": 300
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.046104195481788846,
|
34 |
+
"grad_norm": 0.3389110565185547,
|
35 |
+
"learning_rate": 8.000000000000001e-06,
|
36 |
+
"loss": 2.3696,
|
37 |
+
"step": 400
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.05763024435223606,
|
41 |
+
"grad_norm": 0.38851398229599,
|
42 |
+
"learning_rate": 1e-05,
|
43 |
+
"loss": 2.2957,
|
44 |
+
"step": 500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.06915629322268327,
|
48 |
+
"grad_norm": 0.4800083041191101,
|
49 |
+
"learning_rate": 1.2e-05,
|
50 |
+
"loss": 2.2861,
|
51 |
+
"step": 600
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.08068234209313048,
|
55 |
+
"grad_norm": 0.6024179458618164,
|
56 |
+
"learning_rate": 1.4e-05,
|
57 |
+
"loss": 2.2327,
|
58 |
+
"step": 700
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.09220839096357769,
|
62 |
+
"grad_norm": 0.6432300806045532,
|
63 |
+
"learning_rate": 1.6000000000000003e-05,
|
64 |
+
"loss": 2.192,
|
65 |
+
"step": 800
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.1037344398340249,
|
69 |
+
"grad_norm": 0.8140468001365662,
|
70 |
+
"learning_rate": 1.8e-05,
|
71 |
+
"loss": 2.2126,
|
72 |
+
"step": 900
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.11526048870447211,
|
76 |
+
"grad_norm": 0.7425789833068848,
|
77 |
+
"learning_rate": 2e-05,
|
78 |
+
"loss": 2.1765,
|
79 |
+
"step": 1000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.1267865375749193,
|
83 |
+
"grad_norm": 0.8718421459197998,
|
84 |
+
"learning_rate": 1.999162587972224e-05,
|
85 |
+
"loss": 2.1527,
|
86 |
+
"step": 1100
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.13831258644536654,
|
90 |
+
"grad_norm": 0.8124021887779236,
|
91 |
+
"learning_rate": 1.9966517544067045e-05,
|
92 |
+
"loss": 2.1203,
|
93 |
+
"step": 1200
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.14983863531581373,
|
97 |
+
"grad_norm": 0.7131963968276978,
|
98 |
+
"learning_rate": 1.9924717045078964e-05,
|
99 |
+
"loss": 2.0871,
|
100 |
+
"step": 1300
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.16136468418626096,
|
104 |
+
"grad_norm": 0.8409954309463501,
|
105 |
+
"learning_rate": 1.986629439123924e-05,
|
106 |
+
"loss": 2.1075,
|
107 |
+
"step": 1400
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.17289073305670816,
|
111 |
+
"grad_norm": 0.8746618628501892,
|
112 |
+
"learning_rate": 1.9791347430213906e-05,
|
113 |
+
"loss": 2.0631,
|
114 |
+
"step": 1500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.18441678192715538,
|
118 |
+
"grad_norm": 0.8083209991455078,
|
119 |
+
"learning_rate": 1.970000168497618e-05,
|
120 |
+
"loss": 2.0434,
|
121 |
+
"step": 1600
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.19594283079760258,
|
125 |
+
"grad_norm": 0.7458757162094116,
|
126 |
+
"learning_rate": 1.9592410143577566e-05,
|
127 |
+
"loss": 2.0568,
|
128 |
+
"step": 1700
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.2074688796680498,
|
132 |
+
"grad_norm": 0.9946633577346802,
|
133 |
+
"learning_rate": 1.9468753002919764e-05,
|
134 |
+
"loss": 2.0231,
|
135 |
+
"step": 1800
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.218994928538497,
|
139 |
+
"grad_norm": 0.8859148025512695,
|
140 |
+
"learning_rate": 1.932923736695659e-05,
|
141 |
+
"loss": 2.015,
|
142 |
+
"step": 1900
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.23052097740894423,
|
146 |
+
"grad_norm": 0.7294831871986389,
|
147 |
+
"learning_rate": 1.917409689983128e-05,
|
148 |
+
"loss": 2.0288,
|
149 |
+
"step": 2000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.24204702627939143,
|
153 |
+
"grad_norm": 1.2180194854736328,
|
154 |
+
"learning_rate": 1.900359143453017e-05,
|
155 |
+
"loss": 1.9958,
|
156 |
+
"step": 2100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.2535730751498386,
|
160 |
+
"grad_norm": 0.8870425820350647,
|
161 |
+
"learning_rate": 1.8818006537708148e-05,
|
162 |
+
"loss": 2.0638,
|
163 |
+
"step": 2200
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.26509912402028585,
|
167 |
+
"grad_norm": 0.9545063376426697,
|
168 |
+
"learning_rate": 1.8617653031414752e-05,
|
169 |
+
"loss": 1.9911,
|
170 |
+
"step": 2300
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.2766251728907331,
|
174 |
+
"grad_norm": 1.0659151077270508,
|
175 |
+
"learning_rate": 1.8402866472521944e-05,
|
176 |
+
"loss": 1.9976,
|
177 |
+
"step": 2400
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.28815122176118024,
|
181 |
+
"grad_norm": 1.2996395826339722,
|
182 |
+
"learning_rate": 1.8174006590725364e-05,
|
183 |
+
"loss": 2.0412,
|
184 |
+
"step": 2500
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.29967727063162747,
|
188 |
+
"grad_norm": 1.068005919456482,
|
189 |
+
"learning_rate": 1.7931456686060396e-05,
|
190 |
+
"loss": 2.0828,
|
191 |
+
"step": 2600
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.3112033195020747,
|
195 |
+
"grad_norm": 0.814568281173706,
|
196 |
+
"learning_rate": 1.7675622986942045e-05,
|
197 |
+
"loss": 2.0401,
|
198 |
+
"step": 2700
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.3227293683725219,
|
202 |
+
"grad_norm": 1.509273648262024,
|
203 |
+
"learning_rate": 1.7406933969803815e-05,
|
204 |
+
"loss": 1.99,
|
205 |
+
"step": 2800
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.3342554172429691,
|
209 |
+
"grad_norm": 1.2622814178466797,
|
210 |
+
"learning_rate": 1.7125839641475074e-05,
|
211 |
+
"loss": 2.01,
|
212 |
+
"step": 2900
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.3457814661134163,
|
216 |
+
"grad_norm": 0.9653475284576416,
|
217 |
+
"learning_rate": 1.683281078549878e-05,
|
218 |
+
"loss": 2.0083,
|
219 |
+
"step": 3000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.35730751498386354,
|
223 |
+
"grad_norm": 1.0271968841552734,
|
224 |
+
"learning_rate": 1.6528338173651898e-05,
|
225 |
+
"loss": 1.9546,
|
226 |
+
"step": 3100
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.36883356385431076,
|
230 |
+
"grad_norm": 0.8832419514656067,
|
231 |
+
"learning_rate": 1.6212931743989004e-05,
|
232 |
+
"loss": 1.9482,
|
233 |
+
"step": 3200
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.38035961272475793,
|
237 |
+
"grad_norm": 1.2330228090286255,
|
238 |
+
"learning_rate": 1.5887119746785776e-05,
|
239 |
+
"loss": 1.9766,
|
240 |
+
"step": 3300
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.39188566159520516,
|
244 |
+
"grad_norm": 0.9083070158958435,
|
245 |
+
"learning_rate": 1.5551447859812716e-05,
|
246 |
+
"loss": 1.9726,
|
247 |
+
"step": 3400
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.4034117104656524,
|
251 |
+
"grad_norm": 0.9940272569656372,
|
252 |
+
"learning_rate": 1.5206478274420895e-05,
|
253 |
+
"loss": 1.9762,
|
254 |
+
"step": 3500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.4149377593360996,
|
258 |
+
"grad_norm": 0.9907180070877075,
|
259 |
+
"learning_rate": 1.4852788753970369e-05,
|
260 |
+
"loss": 1.9756,
|
261 |
+
"step": 3600
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.4264638082065468,
|
265 |
+
"grad_norm": 0.9995633363723755,
|
266 |
+
"learning_rate": 1.4490971666178177e-05,
|
267 |
+
"loss": 2.032,
|
268 |
+
"step": 3700
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.437989857076994,
|
272 |
+
"grad_norm": 1.0873900651931763,
|
273 |
+
"learning_rate": 1.412163299100667e-05,
|
274 |
+
"loss": 1.9659,
|
275 |
+
"step": 3800
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.44951590594744123,
|
279 |
+
"grad_norm": 0.8924662470817566,
|
280 |
+
"learning_rate": 1.3745391305753669e-05,
|
281 |
+
"loss": 1.989,
|
282 |
+
"step": 3900
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.46104195481788846,
|
286 |
+
"grad_norm": 0.8575798869132996,
|
287 |
+
"learning_rate": 1.3362876749044334e-05,
|
288 |
+
"loss": 2.0116,
|
289 |
+
"step": 4000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.4725680036883356,
|
293 |
+
"grad_norm": 1.1529802083969116,
|
294 |
+
"learning_rate": 1.2974729965459843e-05,
|
295 |
+
"loss": 1.9598,
|
296 |
+
"step": 4100
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.48409405255878285,
|
300 |
+
"grad_norm": 1.0156272649765015,
|
301 |
+
"learning_rate": 1.2581601032570427e-05,
|
302 |
+
"loss": 1.9568,
|
303 |
+
"step": 4200
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.4956201014292301,
|
307 |
+
"grad_norm": 0.9686967134475708,
|
308 |
+
"learning_rate": 1.2184148372169826e-05,
|
309 |
+
"loss": 1.9608,
|
310 |
+
"step": 4300
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.5071461502996772,
|
314 |
+
"grad_norm": 0.9760933518409729,
|
315 |
+
"learning_rate": 1.1783037647534621e-05,
|
316 |
+
"loss": 1.9117,
|
317 |
+
"step": 4400
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.5186721991701245,
|
321 |
+
"grad_norm": 1.3244590759277344,
|
322 |
+
"learning_rate": 1.1378940648555367e-05,
|
323 |
+
"loss": 1.9626,
|
324 |
+
"step": 4500
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.5301982480405717,
|
328 |
+
"grad_norm": 1.1514652967453003,
|
329 |
+
"learning_rate": 1.0972534166606734e-05,
|
330 |
+
"loss": 1.9407,
|
331 |
+
"step": 4600
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.5417242969110189,
|
335 |
+
"grad_norm": 1.070824384689331,
|
336 |
+
"learning_rate": 1.0564498861041025e-05,
|
337 |
+
"loss": 1.9533,
|
338 |
+
"step": 4700
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.5532503457814661,
|
342 |
+
"grad_norm": 0.9388185739517212,
|
343 |
+
"learning_rate": 1.0155518119203511e-05,
|
344 |
+
"loss": 1.971,
|
345 |
+
"step": 4800
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.5647763946519133,
|
349 |
+
"grad_norm": 0.9750696420669556,
|
350 |
+
"learning_rate": 9.74627691187888e-06,
|
351 |
+
"loss": 1.9924,
|
352 |
+
"step": 4900
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.5763024435223605,
|
356 |
+
"grad_norm": 1.0221482515335083,
|
357 |
+
"learning_rate": 9.337460646085684e-06,
|
358 |
+
"loss": 1.9991,
|
359 |
+
"step": 5000
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.5878284923928078,
|
363 |
+
"grad_norm": 1.3475196361541748,
|
364 |
+
"learning_rate": 8.929754017140172e-06,
|
365 |
+
"loss": 1.9161,
|
366 |
+
"step": 5100
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.5993545412632549,
|
370 |
+
"grad_norm": 1.2582361698150635,
|
371 |
+
"learning_rate": 8.523839861912114e-06,
|
372 |
+
"loss": 1.9163,
|
373 |
+
"step": 5200
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.6108805901337022,
|
377 |
+
"grad_norm": 1.0771867036819458,
|
378 |
+
"learning_rate": 8.120398015193171e-06,
|
379 |
+
"loss": 1.9663,
|
380 |
+
"step": 5300
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.6224066390041494,
|
384 |
+
"grad_norm": 1.1660715341567993,
|
385 |
+
"learning_rate": 7.720104171093247e-06,
|
386 |
+
"loss": 1.956,
|
387 |
+
"step": 5400
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.6339326878745966,
|
391 |
+
"grad_norm": 1.200835943222046,
|
392 |
+
"learning_rate": 7.3236287513717275e-06,
|
393 |
+
"loss": 1.9511,
|
394 |
+
"step": 5500
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.6454587367450438,
|
398 |
+
"grad_norm": 0.9402092099189758,
|
399 |
+
"learning_rate": 6.931635782599e-06,
|
400 |
+
"loss": 1.9615,
|
401 |
+
"step": 5600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.656984785615491,
|
405 |
+
"grad_norm": 0.8676489591598511,
|
406 |
+
"learning_rate": 6.544781784028769e-06,
|
407 |
+
"loss": 1.9899,
|
408 |
+
"step": 5700
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.6685108344859382,
|
412 |
+
"grad_norm": 0.9059965014457703,
|
413 |
+
"learning_rate": 6.163714668043834e-06,
|
414 |
+
"loss": 1.9786,
|
415 |
+
"step": 5800
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.6800368833563855,
|
419 |
+
"grad_norm": 1.3843269348144531,
|
420 |
+
"learning_rate": 5.789072655016817e-06,
|
421 |
+
"loss": 1.9591,
|
422 |
+
"step": 5900
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.6915629322268326,
|
426 |
+
"grad_norm": 1.0491424798965454,
|
427 |
+
"learning_rate": 5.421483204403355e-06,
|
428 |
+
"loss": 1.9437,
|
429 |
+
"step": 6000
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.7030889810972798,
|
433 |
+
"grad_norm": 0.8120349049568176,
|
434 |
+
"learning_rate": 5.061561963857914e-06,
|
435 |
+
"loss": 1.9221,
|
436 |
+
"step": 6100
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.7146150299677271,
|
440 |
+
"grad_norm": 0.9938396215438843,
|
441 |
+
"learning_rate": 4.709911738132252e-06,
|
442 |
+
"loss": 1.982,
|
443 |
+
"step": 6200
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.7261410788381742,
|
447 |
+
"grad_norm": 1.0250656604766846,
|
448 |
+
"learning_rate": 4.367121479483565e-06,
|
449 |
+
"loss": 1.9804,
|
450 |
+
"step": 6300
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.7376671277086215,
|
454 |
+
"grad_norm": 0.9957606792449951,
|
455 |
+
"learning_rate": 4.0337653012830365e-06,
|
456 |
+
"loss": 1.9699,
|
457 |
+
"step": 6400
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.7491931765790687,
|
461 |
+
"grad_norm": 1.3013089895248413,
|
462 |
+
"learning_rate": 3.710401516476997e-06,
|
463 |
+
"loss": 1.9395,
|
464 |
+
"step": 6500
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.7607192254495159,
|
468 |
+
"grad_norm": 0.9715898633003235,
|
469 |
+
"learning_rate": 3.397571702510918e-06,
|
470 |
+
"loss": 1.9448,
|
471 |
+
"step": 6600
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.7722452743199631,
|
475 |
+
"grad_norm": 0.9100465774536133,
|
476 |
+
"learning_rate": 3.0957997942825337e-06,
|
477 |
+
"loss": 1.9169,
|
478 |
+
"step": 6700
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.7837713231904103,
|
482 |
+
"grad_norm": 0.9665968418121338,
|
483 |
+
"learning_rate": 2.8055912066430314e-06,
|
484 |
+
"loss": 1.9538,
|
485 |
+
"step": 6800
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.7952973720608575,
|
489 |
+
"grad_norm": 1.5346043109893799,
|
490 |
+
"learning_rate": 2.52743198791612e-06,
|
491 |
+
"loss": 1.892,
|
492 |
+
"step": 6900
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.8068234209313048,
|
496 |
+
"grad_norm": 0.8448971509933472,
|
497 |
+
"learning_rate": 2.2617880058525942e-06,
|
498 |
+
"loss": 1.9466,
|
499 |
+
"step": 7000
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.8183494698017519,
|
503 |
+
"grad_norm": 1.1289533376693726,
|
504 |
+
"learning_rate": 2.009104167383833e-06,
|
505 |
+
"loss": 1.9142,
|
506 |
+
"step": 7100
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.8298755186721992,
|
510 |
+
"grad_norm": 0.9741213917732239,
|
511 |
+
"learning_rate": 1.769803673480942e-06,
|
512 |
+
"loss": 1.9677,
|
513 |
+
"step": 7200
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.8414015675426464,
|
517 |
+
"grad_norm": 1.1484721899032593,
|
518 |
+
"learning_rate": 1.544287310367625e-06,
|
519 |
+
"loss": 1.9834,
|
520 |
+
"step": 7300
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.8529276164130936,
|
524 |
+
"grad_norm": 1.4257473945617676,
|
525 |
+
"learning_rate": 1.332932778273739e-06,
|
526 |
+
"loss": 1.9686,
|
527 |
+
"step": 7400
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.8644536652835408,
|
531 |
+
"grad_norm": 1.19083833694458,
|
532 |
+
"learning_rate": 1.1360940588538883e-06,
|
533 |
+
"loss": 1.9588,
|
534 |
+
"step": 7500
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.875979714153988,
|
538 |
+
"grad_norm": 1.4401795864105225,
|
539 |
+
"learning_rate": 9.541008223304181e-07,
|
540 |
+
"loss": 1.9146,
|
541 |
+
"step": 7600
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.8875057630244352,
|
545 |
+
"grad_norm": 0.9895892143249512,
|
546 |
+
"learning_rate": 7.872578753538107e-07,
|
547 |
+
"loss": 1.9502,
|
548 |
+
"step": 7700
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.8990318118948825,
|
552 |
+
"grad_norm": 1.0904405117034912,
|
553 |
+
"learning_rate": 6.358446505051574e-07,
|
554 |
+
"loss": 1.947,
|
555 |
+
"step": 7800
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.9105578607653296,
|
559 |
+
"grad_norm": 1.1551321744918823,
|
560 |
+
"learning_rate": 5.001147382957639e-07,
|
561 |
+
"loss": 1.9828,
|
562 |
+
"step": 7900
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.9220839096357769,
|
566 |
+
"grad_norm": 1.3340507745742798,
|
567 |
+
"learning_rate": 3.802954624476585e-07,
|
568 |
+
"loss": 1.8885,
|
569 |
+
"step": 8000
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.9336099585062241,
|
573 |
+
"grad_norm": 0.9464887380599976,
|
574 |
+
"learning_rate": 2.7658749916634864e-07,
|
575 |
+
"loss": 1.9799,
|
576 |
+
"step": 8100
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.9451360073766712,
|
580 |
+
"grad_norm": 1.0508813858032227,
|
581 |
+
"learning_rate": 1.8916454104349058e-07,
|
582 |
+
"loss": 1.9436,
|
583 |
+
"step": 8200
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.9566620562471185,
|
587 |
+
"grad_norm": 1.4017341136932373,
|
588 |
+
"learning_rate": 1.1817300615235672e-07,
|
589 |
+
"loss": 1.9354,
|
590 |
+
"step": 8300
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.9681881051175657,
|
594 |
+
"grad_norm": 0.9612733125686646,
|
595 |
+
"learning_rate": 6.373179282332275e-08,
|
596 |
+
"loss": 1.9696,
|
597 |
+
"step": 8400
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.9797141539880129,
|
601 |
+
"grad_norm": 1.061047911643982,
|
602 |
+
"learning_rate": 2.6227471163585705e-08,
|
603 |
+
"loss": 1.9373,
|
604 |
+
"step": 8500
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.9912402028584602,
|
608 |
+
"grad_norm": 1.4511463642120361,
|
609 |
+
"learning_rate": 4.965298070647162e-09,
|
610 |
+
"loss": 1.925,
|
611 |
+
"step": 8600
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 1.0,
|
615 |
+
"step": 8676,
|
616 |
+
"total_flos": 1.58472345255936e+17,
|
617 |
+
"train_loss": 2.0172890398896968,
|
618 |
+
"train_runtime": 2382.4555,
|
619 |
+
"train_samples_per_second": 7.283,
|
620 |
+
"train_steps_per_second": 3.642
|
621 |
+
}
|
622 |
+
],
|
623 |
+
"logging_steps": 100,
|
624 |
+
"max_steps": 8676,
|
625 |
+
"num_input_tokens_seen": 0,
|
626 |
+
"num_train_epochs": 1,
|
627 |
+
"save_steps": 500,
|
628 |
+
"stateful_callbacks": {
|
629 |
+
"TrainerControl": {
|
630 |
+
"args": {
|
631 |
+
"should_epoch_stop": false,
|
632 |
+
"should_evaluate": false,
|
633 |
+
"should_log": false,
|
634 |
+
"should_save": true,
|
635 |
+
"should_training_stop": true
|
636 |
+
},
|
637 |
+
"attributes": {}
|
638 |
+
}
|
639 |
+
},
|
640 |
+
"total_flos": 1.58472345255936e+17,
|
641 |
+
"train_batch_size": 2,
|
642 |
+
"trial_name": null,
|
643 |
+
"trial_params": null
|
644 |
+
}
|