MHGanainy/gpt2-xl-lora-multi-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 +665 -0
README.md
CHANGED
@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
# gpt2-xl-lora-multi-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-multi-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: 2.0126
|
20 |
|
21 |
## Model description
|
22 |
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_loss": 2.0126099586486816,
|
4 |
+
"eval_runtime": 4087.7012,
|
5 |
+
"eval_samples_per_second": 7.809,
|
6 |
+
"eval_steps_per_second": 0.488,
|
7 |
+
"perplexity": 7.482821738733412,
|
8 |
+
"total_flos": 1.3078189256852111e+18,
|
9 |
+
"train_loss": 2.2911263094506222,
|
10 |
+
"train_runtime": 2843.733,
|
11 |
+
"train_samples_per_second": 50.495,
|
12 |
+
"train_steps_per_second": 3.156
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_loss": 2.0126099586486816,
|
4 |
+
"eval_runtime": 4087.7012,
|
5 |
+
"eval_samples_per_second": 7.809,
|
6 |
+
"eval_steps_per_second": 0.488,
|
7 |
+
"perplexity": 7.482821738733412
|
8 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"total_flos": 1.3078189256852111e+18,
|
4 |
+
"train_loss": 2.2911263094506222,
|
5 |
+
"train_runtime": 2843.733,
|
6 |
+
"train_samples_per_second": 50.495,
|
7 |
+
"train_steps_per_second": 3.156
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,665 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 8975,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.011142061281337047,
|
13 |
+
"grad_norm": 0.04801425710320473,
|
14 |
+
"learning_rate": 2.7859033291544785e-07,
|
15 |
+
"loss": 2.4677,
|
16 |
+
"step": 100
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.022284122562674095,
|
20 |
+
"grad_norm": 0.04515479877591133,
|
21 |
+
"learning_rate": 5.571806658308957e-07,
|
22 |
+
"loss": 2.465,
|
23 |
+
"step": 200
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.033426183844011144,
|
27 |
+
"grad_norm": 0.04905751347541809,
|
28 |
+
"learning_rate": 8.357709987463436e-07,
|
29 |
+
"loss": 2.479,
|
30 |
+
"step": 300
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.04456824512534819,
|
34 |
+
"grad_norm": 0.0585104376077652,
|
35 |
+
"learning_rate": 1.1143613316617914e-06,
|
36 |
+
"loss": 2.4715,
|
37 |
+
"step": 400
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.055710306406685235,
|
41 |
+
"grad_norm": 0.06588415056467056,
|
42 |
+
"learning_rate": 1.3929516645772392e-06,
|
43 |
+
"loss": 2.4912,
|
44 |
+
"step": 500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.06685236768802229,
|
48 |
+
"grad_norm": 0.07128334790468216,
|
49 |
+
"learning_rate": 1.6715419974926873e-06,
|
50 |
+
"loss": 2.449,
|
51 |
+
"step": 600
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.07799442896935933,
|
55 |
+
"grad_norm": 0.08448737859725952,
|
56 |
+
"learning_rate": 1.9501323304081347e-06,
|
57 |
+
"loss": 2.4613,
|
58 |
+
"step": 700
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.08913649025069638,
|
62 |
+
"grad_norm": 0.09213274717330933,
|
63 |
+
"learning_rate": 2.2287226633235828e-06,
|
64 |
+
"loss": 2.4366,
|
65 |
+
"step": 800
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.10027855153203342,
|
69 |
+
"grad_norm": 0.10825659334659576,
|
70 |
+
"learning_rate": 2.507312996239031e-06,
|
71 |
+
"loss": 2.4342,
|
72 |
+
"step": 900
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.11142061281337047,
|
76 |
+
"grad_norm": 0.11989778280258179,
|
77 |
+
"learning_rate": 2.7859033291544785e-06,
|
78 |
+
"loss": 2.4227,
|
79 |
+
"step": 1000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.12256267409470752,
|
83 |
+
"grad_norm": 0.11828930675983429,
|
84 |
+
"learning_rate": 3.064493662069926e-06,
|
85 |
+
"loss": 2.4346,
|
86 |
+
"step": 1100
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.13370473537604458,
|
90 |
+
"grad_norm": 0.13713237643241882,
|
91 |
+
"learning_rate": 3.3430839949853746e-06,
|
92 |
+
"loss": 2.3954,
|
93 |
+
"step": 1200
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.14484679665738162,
|
97 |
+
"grad_norm": 0.13783149421215057,
|
98 |
+
"learning_rate": 3.6216743279008222e-06,
|
99 |
+
"loss": 2.4027,
|
100 |
+
"step": 1300
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.15598885793871867,
|
104 |
+
"grad_norm": 0.14380280673503876,
|
105 |
+
"learning_rate": 3.9002646608162694e-06,
|
106 |
+
"loss": 2.3894,
|
107 |
+
"step": 1400
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.1671309192200557,
|
111 |
+
"grad_norm": 0.1586257964372635,
|
112 |
+
"learning_rate": 4.178854993731718e-06,
|
113 |
+
"loss": 2.3923,
|
114 |
+
"step": 1500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.17827298050139276,
|
118 |
+
"grad_norm": 0.15052290260791779,
|
119 |
+
"learning_rate": 4.4574453266471655e-06,
|
120 |
+
"loss": 2.3792,
|
121 |
+
"step": 1600
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.1894150417827298,
|
125 |
+
"grad_norm": 0.16659782826900482,
|
126 |
+
"learning_rate": 4.736035659562614e-06,
|
127 |
+
"loss": 2.3657,
|
128 |
+
"step": 1700
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.20055710306406685,
|
132 |
+
"grad_norm": 0.16844528913497925,
|
133 |
+
"learning_rate": 5.014625992478062e-06,
|
134 |
+
"loss": 2.3748,
|
135 |
+
"step": 1800
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.2116991643454039,
|
139 |
+
"grad_norm": 0.17568887770175934,
|
140 |
+
"learning_rate": 5.29321632539351e-06,
|
141 |
+
"loss": 2.3609,
|
142 |
+
"step": 1900
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.22284122562674094,
|
146 |
+
"grad_norm": 0.19198767840862274,
|
147 |
+
"learning_rate": 5.571806658308957e-06,
|
148 |
+
"loss": 2.3544,
|
149 |
+
"step": 2000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.233983286908078,
|
153 |
+
"grad_norm": 0.19439826905727386,
|
154 |
+
"learning_rate": 5.850396991224405e-06,
|
155 |
+
"loss": 2.3589,
|
156 |
+
"step": 2100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.24512534818941503,
|
160 |
+
"grad_norm": 0.2022334635257721,
|
161 |
+
"learning_rate": 6.128987324139852e-06,
|
162 |
+
"loss": 2.3612,
|
163 |
+
"step": 2200
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.2562674094707521,
|
167 |
+
"grad_norm": 0.20067375898361206,
|
168 |
+
"learning_rate": 6.4075776570553e-06,
|
169 |
+
"loss": 2.3417,
|
170 |
+
"step": 2300
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.26740947075208915,
|
174 |
+
"grad_norm": 0.2165047824382782,
|
175 |
+
"learning_rate": 6.686167989970749e-06,
|
176 |
+
"loss": 2.3481,
|
177 |
+
"step": 2400
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.2785515320334262,
|
181 |
+
"grad_norm": 0.21726970374584198,
|
182 |
+
"learning_rate": 6.9647583228861955e-06,
|
183 |
+
"loss": 2.3331,
|
184 |
+
"step": 2500
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.28969359331476324,
|
188 |
+
"grad_norm": 0.21808430552482605,
|
189 |
+
"learning_rate": 7.2433486558016444e-06,
|
190 |
+
"loss": 2.336,
|
191 |
+
"step": 2600
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.3008356545961003,
|
195 |
+
"grad_norm": 0.23211392760276794,
|
196 |
+
"learning_rate": 7.5219389887170925e-06,
|
197 |
+
"loss": 2.3332,
|
198 |
+
"step": 2700
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.31197771587743733,
|
202 |
+
"grad_norm": 0.24134476482868195,
|
203 |
+
"learning_rate": 7.800529321632539e-06,
|
204 |
+
"loss": 2.3377,
|
205 |
+
"step": 2800
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.3231197771587744,
|
209 |
+
"grad_norm": 0.2478109449148178,
|
210 |
+
"learning_rate": 8.079119654547987e-06,
|
211 |
+
"loss": 2.3209,
|
212 |
+
"step": 2900
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.3342618384401114,
|
216 |
+
"grad_norm": 0.25040534138679504,
|
217 |
+
"learning_rate": 8.357709987463437e-06,
|
218 |
+
"loss": 2.3218,
|
219 |
+
"step": 3000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.34540389972144847,
|
223 |
+
"grad_norm": 0.25530460476875305,
|
224 |
+
"learning_rate": 8.636300320378883e-06,
|
225 |
+
"loss": 2.3063,
|
226 |
+
"step": 3100
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.3565459610027855,
|
230 |
+
"grad_norm": 0.258489727973938,
|
231 |
+
"learning_rate": 8.914890653294331e-06,
|
232 |
+
"loss": 2.3218,
|
233 |
+
"step": 3200
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.36768802228412256,
|
237 |
+
"grad_norm": 0.28399255871772766,
|
238 |
+
"learning_rate": 9.193480986209779e-06,
|
239 |
+
"loss": 2.3063,
|
240 |
+
"step": 3300
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.3788300835654596,
|
244 |
+
"grad_norm": 0.2863540053367615,
|
245 |
+
"learning_rate": 9.472071319125227e-06,
|
246 |
+
"loss": 2.2997,
|
247 |
+
"step": 3400
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.38997214484679665,
|
251 |
+
"grad_norm": 0.2868782877922058,
|
252 |
+
"learning_rate": 9.750661652040675e-06,
|
253 |
+
"loss": 2.2947,
|
254 |
+
"step": 3500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.4011142061281337,
|
258 |
+
"grad_norm": 0.2859346270561218,
|
259 |
+
"learning_rate": 1.0029251984956123e-05,
|
260 |
+
"loss": 2.3054,
|
261 |
+
"step": 3600
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.41225626740947074,
|
265 |
+
"grad_norm": 0.29307645559310913,
|
266 |
+
"learning_rate": 1.030784231787157e-05,
|
267 |
+
"loss": 2.2882,
|
268 |
+
"step": 3700
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.4233983286908078,
|
272 |
+
"grad_norm": 0.3006058633327484,
|
273 |
+
"learning_rate": 1.058643265078702e-05,
|
274 |
+
"loss": 2.2893,
|
275 |
+
"step": 3800
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.43454038997214484,
|
279 |
+
"grad_norm": 0.31586816906929016,
|
280 |
+
"learning_rate": 1.0865022983702467e-05,
|
281 |
+
"loss": 2.2967,
|
282 |
+
"step": 3900
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.4456824512534819,
|
286 |
+
"grad_norm": 0.3375140428543091,
|
287 |
+
"learning_rate": 1.1143613316617914e-05,
|
288 |
+
"loss": 2.2694,
|
289 |
+
"step": 4000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.4568245125348189,
|
293 |
+
"grad_norm": 0.30628982186317444,
|
294 |
+
"learning_rate": 1.142220364953336e-05,
|
295 |
+
"loss": 2.2709,
|
296 |
+
"step": 4100
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.467966573816156,
|
300 |
+
"grad_norm": 0.34296101331710815,
|
301 |
+
"learning_rate": 1.170079398244881e-05,
|
302 |
+
"loss": 2.2783,
|
303 |
+
"step": 4200
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.479108635097493,
|
307 |
+
"grad_norm": 0.33544307947158813,
|
308 |
+
"learning_rate": 1.1979384315364258e-05,
|
309 |
+
"loss": 2.2825,
|
310 |
+
"step": 4300
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.49025069637883006,
|
314 |
+
"grad_norm": 0.3508812487125397,
|
315 |
+
"learning_rate": 1.2257974648279704e-05,
|
316 |
+
"loss": 2.2793,
|
317 |
+
"step": 4400
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.5013927576601671,
|
321 |
+
"grad_norm": 0.34851330518722534,
|
322 |
+
"learning_rate": 1.2536564981195154e-05,
|
323 |
+
"loss": 2.2703,
|
324 |
+
"step": 4500
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.5125348189415042,
|
328 |
+
"grad_norm": 0.3589787185192108,
|
329 |
+
"learning_rate": 1.28151553141106e-05,
|
330 |
+
"loss": 2.2649,
|
331 |
+
"step": 4600
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.5236768802228412,
|
335 |
+
"grad_norm": 0.3842375576496124,
|
336 |
+
"learning_rate": 1.3093745647026049e-05,
|
337 |
+
"loss": 2.264,
|
338 |
+
"step": 4700
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.5348189415041783,
|
342 |
+
"grad_norm": 0.3963667154312134,
|
343 |
+
"learning_rate": 1.3372335979941498e-05,
|
344 |
+
"loss": 2.2567,
|
345 |
+
"step": 4800
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.5459610027855153,
|
349 |
+
"grad_norm": 0.40061014890670776,
|
350 |
+
"learning_rate": 1.3650926312856945e-05,
|
351 |
+
"loss": 2.2567,
|
352 |
+
"step": 4900
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.5571030640668524,
|
356 |
+
"grad_norm": 0.3819388747215271,
|
357 |
+
"learning_rate": 1.3929516645772391e-05,
|
358 |
+
"loss": 2.2545,
|
359 |
+
"step": 5000
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.5682451253481894,
|
363 |
+
"grad_norm": 0.40177789330482483,
|
364 |
+
"learning_rate": 1.420810697868784e-05,
|
365 |
+
"loss": 2.2509,
|
366 |
+
"step": 5100
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.5793871866295265,
|
370 |
+
"grad_norm": 0.4258297085762024,
|
371 |
+
"learning_rate": 1.4486697311603289e-05,
|
372 |
+
"loss": 2.2466,
|
373 |
+
"step": 5200
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.5905292479108635,
|
377 |
+
"grad_norm": 0.44000041484832764,
|
378 |
+
"learning_rate": 1.4765287644518735e-05,
|
379 |
+
"loss": 2.2453,
|
380 |
+
"step": 5300
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.6016713091922006,
|
384 |
+
"grad_norm": 0.4116784632205963,
|
385 |
+
"learning_rate": 1.5043877977434185e-05,
|
386 |
+
"loss": 2.2708,
|
387 |
+
"step": 5400
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.6128133704735376,
|
391 |
+
"grad_norm": 0.39489638805389404,
|
392 |
+
"learning_rate": 1.5322468310349633e-05,
|
393 |
+
"loss": 2.2656,
|
394 |
+
"step": 5500
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.6239554317548747,
|
398 |
+
"grad_norm": 0.43062567710876465,
|
399 |
+
"learning_rate": 1.5601058643265078e-05,
|
400 |
+
"loss": 2.2305,
|
401 |
+
"step": 5600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.6350974930362117,
|
405 |
+
"grad_norm": 0.4567050635814667,
|
406 |
+
"learning_rate": 1.587964897618053e-05,
|
407 |
+
"loss": 2.2392,
|
408 |
+
"step": 5700
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.6462395543175488,
|
412 |
+
"grad_norm": 0.40272602438926697,
|
413 |
+
"learning_rate": 1.6158239309095974e-05,
|
414 |
+
"loss": 2.2327,
|
415 |
+
"step": 5800
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.6573816155988857,
|
419 |
+
"grad_norm": 0.4182547628879547,
|
420 |
+
"learning_rate": 1.6436829642011422e-05,
|
421 |
+
"loss": 2.227,
|
422 |
+
"step": 5900
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.6685236768802229,
|
426 |
+
"grad_norm": 0.4616607129573822,
|
427 |
+
"learning_rate": 1.6715419974926873e-05,
|
428 |
+
"loss": 2.223,
|
429 |
+
"step": 6000
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.6796657381615598,
|
433 |
+
"grad_norm": 0.4605095386505127,
|
434 |
+
"learning_rate": 1.6994010307842318e-05,
|
435 |
+
"loss": 2.229,
|
436 |
+
"step": 6100
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.6908077994428969,
|
440 |
+
"grad_norm": 0.48064395785331726,
|
441 |
+
"learning_rate": 1.7272600640757766e-05,
|
442 |
+
"loss": 2.2285,
|
443 |
+
"step": 6200
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.7019498607242339,
|
447 |
+
"grad_norm": 0.45375484228134155,
|
448 |
+
"learning_rate": 1.7551190973673214e-05,
|
449 |
+
"loss": 2.231,
|
450 |
+
"step": 6300
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.713091922005571,
|
454 |
+
"grad_norm": 0.4386723041534424,
|
455 |
+
"learning_rate": 1.7829781306588662e-05,
|
456 |
+
"loss": 2.2295,
|
457 |
+
"step": 6400
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.724233983286908,
|
461 |
+
"grad_norm": 0.4519370198249817,
|
462 |
+
"learning_rate": 1.810837163950411e-05,
|
463 |
+
"loss": 2.2185,
|
464 |
+
"step": 6500
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.7353760445682451,
|
468 |
+
"grad_norm": 0.45770174264907837,
|
469 |
+
"learning_rate": 1.8386961972419558e-05,
|
470 |
+
"loss": 2.2146,
|
471 |
+
"step": 6600
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.7465181058495822,
|
475 |
+
"grad_norm": 0.46970096230506897,
|
476 |
+
"learning_rate": 1.8665552305335006e-05,
|
477 |
+
"loss": 2.2268,
|
478 |
+
"step": 6700
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.7576601671309192,
|
482 |
+
"grad_norm": 0.4301040768623352,
|
483 |
+
"learning_rate": 1.8944142638250454e-05,
|
484 |
+
"loss": 2.2186,
|
485 |
+
"step": 6800
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.7688022284122563,
|
489 |
+
"grad_norm": 0.49275335669517517,
|
490 |
+
"learning_rate": 1.9222732971165902e-05,
|
491 |
+
"loss": 2.2143,
|
492 |
+
"step": 6900
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.7799442896935933,
|
496 |
+
"grad_norm": 0.4531406760215759,
|
497 |
+
"learning_rate": 1.950132330408135e-05,
|
498 |
+
"loss": 2.2087,
|
499 |
+
"step": 7000
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.7910863509749304,
|
503 |
+
"grad_norm": 0.44058841466903687,
|
504 |
+
"learning_rate": 1.97799136369968e-05,
|
505 |
+
"loss": 2.2081,
|
506 |
+
"step": 7100
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.8022284122562674,
|
510 |
+
"grad_norm": 0.4620118737220764,
|
511 |
+
"learning_rate": 1.9993253992196043e-05,
|
512 |
+
"loss": 2.2029,
|
513 |
+
"step": 7200
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.8133704735376045,
|
517 |
+
"grad_norm": 0.4749448299407959,
|
518 |
+
"learning_rate": 1.9776845355951603e-05,
|
519 |
+
"loss": 2.1974,
|
520 |
+
"step": 7300
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.8245125348189415,
|
524 |
+
"grad_norm": 0.4482609033584595,
|
525 |
+
"learning_rate": 1.9262050899616325e-05,
|
526 |
+
"loss": 2.2229,
|
527 |
+
"step": 7400
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.8356545961002786,
|
531 |
+
"grad_norm": 0.4689159095287323,
|
532 |
+
"learning_rate": 1.8464581965697866e-05,
|
533 |
+
"loss": 2.2081,
|
534 |
+
"step": 7500
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.8467966573816156,
|
538 |
+
"grad_norm": 0.4448583126068115,
|
539 |
+
"learning_rate": 1.7408777020560473e-05,
|
540 |
+
"loss": 2.2181,
|
541 |
+
"step": 7600
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.8579387186629527,
|
545 |
+
"grad_norm": 0.513080894947052,
|
546 |
+
"learning_rate": 1.6126858853144854e-05,
|
547 |
+
"loss": 2.2045,
|
548 |
+
"step": 7700
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.8690807799442897,
|
552 |
+
"grad_norm": 0.4850214421749115,
|
553 |
+
"learning_rate": 1.465795114698568e-05,
|
554 |
+
"loss": 2.1966,
|
555 |
+
"step": 7800
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.8802228412256268,
|
559 |
+
"grad_norm": 0.5757237672805786,
|
560 |
+
"learning_rate": 1.3046884439396632e-05,
|
561 |
+
"loss": 2.2019,
|
562 |
+
"step": 7900
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.8913649025069638,
|
566 |
+
"grad_norm": 0.48911547660827637,
|
567 |
+
"learning_rate": 1.1342827909521198e-05,
|
568 |
+
"loss": 2.1994,
|
569 |
+
"step": 8000
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.9025069637883009,
|
573 |
+
"grad_norm": 0.5143416523933411,
|
574 |
+
"learning_rate": 9.597788752588108e-06,
|
575 |
+
"loss": 2.207,
|
576 |
+
"step": 8100
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.9136490250696379,
|
580 |
+
"grad_norm": 0.483064204454422,
|
581 |
+
"learning_rate": 7.882117054290375e-06,
|
582 |
+
"loss": 2.1971,
|
583 |
+
"step": 8200
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.924791086350975,
|
587 |
+
"grad_norm": 0.4610692262649536,
|
588 |
+
"learning_rate": 6.213603769467132e-06,
|
589 |
+
"loss": 2.1977,
|
590 |
+
"step": 8300
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.935933147632312,
|
594 |
+
"grad_norm": 0.50405353307724,
|
595 |
+
"learning_rate": 4.660649942045826e-06,
|
596 |
+
"loss": 2.1955,
|
597 |
+
"step": 8400
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.947075208913649,
|
601 |
+
"grad_norm": 0.47849225997924805,
|
602 |
+
"learning_rate": 3.27065116687201e-06,
|
603 |
+
"loss": 2.1906,
|
604 |
+
"step": 8500
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.958217270194986,
|
608 |
+
"grad_norm": 0.49040549993515015,
|
609 |
+
"learning_rate": 2.0860297089355943e-06,
|
610 |
+
"loss": 2.1875,
|
611 |
+
"step": 8600
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.9693593314763231,
|
615 |
+
"grad_norm": 0.49369877576828003,
|
616 |
+
"learning_rate": 1.1429397910307794e-06,
|
617 |
+
"loss": 2.2054,
|
618 |
+
"step": 8700
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.9805013927576601,
|
622 |
+
"grad_norm": 0.5104972124099731,
|
623 |
+
"learning_rate": 4.701641798198353e-07,
|
624 |
+
"loss": 2.2121,
|
625 |
+
"step": 8800
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.9916434540389972,
|
629 |
+
"grad_norm": 0.48128950595855713,
|
630 |
+
"learning_rate": 8.823574609897134e-08,
|
631 |
+
"loss": 2.1928,
|
632 |
+
"step": 8900
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 1.0,
|
636 |
+
"step": 8975,
|
637 |
+
"total_flos": 1.3078189256852111e+18,
|
638 |
+
"train_loss": 2.2911263094506222,
|
639 |
+
"train_runtime": 2843.733,
|
640 |
+
"train_samples_per_second": 50.495,
|
641 |
+
"train_steps_per_second": 3.156
|
642 |
+
}
|
643 |
+
],
|
644 |
+
"logging_steps": 100,
|
645 |
+
"max_steps": 8975,
|
646 |
+
"num_input_tokens_seen": 0,
|
647 |
+
"num_train_epochs": 1,
|
648 |
+
"save_steps": 500,
|
649 |
+
"stateful_callbacks": {
|
650 |
+
"TrainerControl": {
|
651 |
+
"args": {
|
652 |
+
"should_epoch_stop": false,
|
653 |
+
"should_evaluate": false,
|
654 |
+
"should_log": false,
|
655 |
+
"should_save": true,
|
656 |
+
"should_training_stop": true
|
657 |
+
},
|
658 |
+
"attributes": {}
|
659 |
+
}
|
660 |
+
},
|
661 |
+
"total_flos": 1.3078189256852111e+18,
|
662 |
+
"train_batch_size": 2,
|
663 |
+
"trial_name": null,
|
664 |
+
"trial_params": null
|
665 |
+
}
|