MHGanainy/gpt2-xl-lora-multi-512-7
Browse files- README.md +2 -0
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Oct20_16-50-59_d0f171ada4b3/events.out.tfevents.1729445336.d0f171ada4b3.6939.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +805 -0
README.md
CHANGED
@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
# gpt2-xl-lora-multi-512-7
|
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-512-7
|
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: 2.2163
|
20 |
|
21 |
## Model description
|
22 |
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_loss": 2.216308832168579,
|
4 |
+
"eval_runtime": 127.3922,
|
5 |
+
"eval_samples_per_second": 246.271,
|
6 |
+
"eval_steps_per_second": 15.393,
|
7 |
+
"perplexity": 9.173407708925824,
|
8 |
+
"total_flos": 1.59124040841796e+18,
|
9 |
+
"train_loss": 2.428147726093893,
|
10 |
+
"train_runtime": 2109.7476,
|
11 |
+
"train_samples_per_second": 82.811,
|
12 |
+
"train_steps_per_second": 5.176
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_loss": 2.216308832168579,
|
4 |
+
"eval_runtime": 127.3922,
|
5 |
+
"eval_samples_per_second": 246.271,
|
6 |
+
"eval_steps_per_second": 15.393,
|
7 |
+
"perplexity": 9.173407708925824
|
8 |
+
}
|
runs/Oct20_16-50-59_d0f171ada4b3/events.out.tfevents.1729445336.d0f171ada4b3.6939.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f1291a62f9248604f40cb5726c7503e46cce268df4669bab47c43480458d5e57
|
3 |
+
size 311
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"total_flos": 1.59124040841796e+18,
|
4 |
+
"train_loss": 2.428147726093893,
|
5 |
+
"train_runtime": 2109.7476,
|
6 |
+
"train_samples_per_second": 82.811,
|
7 |
+
"train_steps_per_second": 5.176
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,805 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 10920,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.009157509157509158,
|
13 |
+
"grad_norm": 0.10490927845239639,
|
14 |
+
"learning_rate": 2.289639381797367e-07,
|
15 |
+
"loss": 2.7059,
|
16 |
+
"step": 100
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.018315018315018316,
|
20 |
+
"grad_norm": 0.22449691593647003,
|
21 |
+
"learning_rate": 4.5563823697767606e-07,
|
22 |
+
"loss": 2.7179,
|
23 |
+
"step": 200
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.027472527472527472,
|
27 |
+
"grad_norm": 0.14629626274108887,
|
28 |
+
"learning_rate": 6.846021751574127e-07,
|
29 |
+
"loss": 2.7183,
|
30 |
+
"step": 300
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.03663003663003663,
|
34 |
+
"grad_norm": 0.19104591012001038,
|
35 |
+
"learning_rate": 9.112764739553521e-07,
|
36 |
+
"loss": 2.715,
|
37 |
+
"step": 400
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.045787545787545784,
|
41 |
+
"grad_norm": 0.2343115508556366,
|
42 |
+
"learning_rate": 1.1402404121350888e-06,
|
43 |
+
"loss": 2.6869,
|
44 |
+
"step": 500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.054945054945054944,
|
48 |
+
"grad_norm": 0.1325504034757614,
|
49 |
+
"learning_rate": 1.366914710933028e-06,
|
50 |
+
"loss": 2.6707,
|
51 |
+
"step": 600
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.0641025641025641,
|
55 |
+
"grad_norm": 0.22846511006355286,
|
56 |
+
"learning_rate": 1.5958786491127649e-06,
|
57 |
+
"loss": 2.6654,
|
58 |
+
"step": 700
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.07326007326007326,
|
62 |
+
"grad_norm": 0.1422765702009201,
|
63 |
+
"learning_rate": 1.8248425872925015e-06,
|
64 |
+
"loss": 2.6269,
|
65 |
+
"step": 800
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.08241758241758242,
|
69 |
+
"grad_norm": 0.33947649598121643,
|
70 |
+
"learning_rate": 2.0538065254722382e-06,
|
71 |
+
"loss": 2.5998,
|
72 |
+
"step": 900
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.09157509157509157,
|
76 |
+
"grad_norm": 0.12349778413772583,
|
77 |
+
"learning_rate": 2.2804808242701776e-06,
|
78 |
+
"loss": 2.586,
|
79 |
+
"step": 1000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.10073260073260074,
|
83 |
+
"grad_norm": 0.21951575577259064,
|
84 |
+
"learning_rate": 2.5094447624499143e-06,
|
85 |
+
"loss": 2.5917,
|
86 |
+
"step": 1100
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.10989010989010989,
|
90 |
+
"grad_norm": 0.16132010519504547,
|
91 |
+
"learning_rate": 2.738408700629651e-06,
|
92 |
+
"loss": 2.5675,
|
93 |
+
"step": 1200
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.11904761904761904,
|
97 |
+
"grad_norm": 0.31237655878067017,
|
98 |
+
"learning_rate": 2.9673726388093876e-06,
|
99 |
+
"loss": 2.5556,
|
100 |
+
"step": 1300
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.1282051282051282,
|
104 |
+
"grad_norm": 0.15914414823055267,
|
105 |
+
"learning_rate": 3.1963365769891247e-06,
|
106 |
+
"loss": 2.5326,
|
107 |
+
"step": 1400
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.13736263736263737,
|
111 |
+
"grad_norm": 0.1860857754945755,
|
112 |
+
"learning_rate": 3.425300515168861e-06,
|
113 |
+
"loss": 2.5353,
|
114 |
+
"step": 1500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.14652014652014653,
|
118 |
+
"grad_norm": 0.18469680845737457,
|
119 |
+
"learning_rate": 3.6542644533485977e-06,
|
120 |
+
"loss": 2.5114,
|
121 |
+
"step": 1600
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.15567765567765568,
|
125 |
+
"grad_norm": 0.3093177080154419,
|
126 |
+
"learning_rate": 3.883228391528335e-06,
|
127 |
+
"loss": 2.5025,
|
128 |
+
"step": 1700
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.16483516483516483,
|
132 |
+
"grad_norm": 0.29793912172317505,
|
133 |
+
"learning_rate": 4.112192329708071e-06,
|
134 |
+
"loss": 2.4835,
|
135 |
+
"step": 1800
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.17399267399267399,
|
139 |
+
"grad_norm": 0.20354346930980682,
|
140 |
+
"learning_rate": 4.341156267887808e-06,
|
141 |
+
"loss": 2.4779,
|
142 |
+
"step": 1900
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.18315018315018314,
|
146 |
+
"grad_norm": 0.20856451988220215,
|
147 |
+
"learning_rate": 4.5701202060675444e-06,
|
148 |
+
"loss": 2.4694,
|
149 |
+
"step": 2000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.19230769230769232,
|
153 |
+
"grad_norm": 0.17660389840602875,
|
154 |
+
"learning_rate": 4.7990841442472815e-06,
|
155 |
+
"loss": 2.4717,
|
156 |
+
"step": 2100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.20146520146520147,
|
160 |
+
"grad_norm": 0.1957291066646576,
|
161 |
+
"learning_rate": 5.028048082427019e-06,
|
162 |
+
"loss": 2.4643,
|
163 |
+
"step": 2200
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.21062271062271062,
|
167 |
+
"grad_norm": 0.201618030667305,
|
168 |
+
"learning_rate": 5.257012020606756e-06,
|
169 |
+
"loss": 2.4694,
|
170 |
+
"step": 2300
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.21978021978021978,
|
174 |
+
"grad_norm": 0.22514258325099945,
|
175 |
+
"learning_rate": 5.485975958786491e-06,
|
176 |
+
"loss": 2.4528,
|
177 |
+
"step": 2400
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.22893772893772893,
|
181 |
+
"grad_norm": 0.2677931785583496,
|
182 |
+
"learning_rate": 5.714939896966228e-06,
|
183 |
+
"loss": 2.454,
|
184 |
+
"step": 2500
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.23809523809523808,
|
188 |
+
"grad_norm": 0.2718426287174225,
|
189 |
+
"learning_rate": 5.943903835145965e-06,
|
190 |
+
"loss": 2.4632,
|
191 |
+
"step": 2600
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.24725274725274726,
|
195 |
+
"grad_norm": 0.2885560691356659,
|
196 |
+
"learning_rate": 6.172867773325702e-06,
|
197 |
+
"loss": 2.4549,
|
198 |
+
"step": 2700
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.2564102564102564,
|
202 |
+
"grad_norm": 0.19180075824260712,
|
203 |
+
"learning_rate": 6.401831711505439e-06,
|
204 |
+
"loss": 2.4496,
|
205 |
+
"step": 2800
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.26556776556776557,
|
209 |
+
"grad_norm": 0.22613513469696045,
|
210 |
+
"learning_rate": 6.630795649685176e-06,
|
211 |
+
"loss": 2.4528,
|
212 |
+
"step": 2900
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.27472527472527475,
|
216 |
+
"grad_norm": 0.19509850442409515,
|
217 |
+
"learning_rate": 6.859759587864911e-06,
|
218 |
+
"loss": 2.4441,
|
219 |
+
"step": 3000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.2838827838827839,
|
223 |
+
"grad_norm": 0.21075837314128876,
|
224 |
+
"learning_rate": 7.088723526044648e-06,
|
225 |
+
"loss": 2.4464,
|
226 |
+
"step": 3100
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.29304029304029305,
|
230 |
+
"grad_norm": 0.2654266357421875,
|
231 |
+
"learning_rate": 7.3176874642243855e-06,
|
232 |
+
"loss": 2.4298,
|
233 |
+
"step": 3200
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.3021978021978022,
|
237 |
+
"grad_norm": 0.27135413885116577,
|
238 |
+
"learning_rate": 7.546651402404122e-06,
|
239 |
+
"loss": 2.4364,
|
240 |
+
"step": 3300
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.31135531135531136,
|
244 |
+
"grad_norm": 0.27136310935020447,
|
245 |
+
"learning_rate": 7.773325701202062e-06,
|
246 |
+
"loss": 2.4287,
|
247 |
+
"step": 3400
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.32051282051282054,
|
251 |
+
"grad_norm": 0.2661312222480774,
|
252 |
+
"learning_rate": 8.002289639381798e-06,
|
253 |
+
"loss": 2.4275,
|
254 |
+
"step": 3500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.32967032967032966,
|
258 |
+
"grad_norm": 0.20097434520721436,
|
259 |
+
"learning_rate": 8.231253577561534e-06,
|
260 |
+
"loss": 2.4336,
|
261 |
+
"step": 3600
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.33882783882783885,
|
265 |
+
"grad_norm": 0.3780891001224518,
|
266 |
+
"learning_rate": 8.46021751574127e-06,
|
267 |
+
"loss": 2.4372,
|
268 |
+
"step": 3700
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.34798534798534797,
|
272 |
+
"grad_norm": 0.23288694024085999,
|
273 |
+
"learning_rate": 8.689181453921009e-06,
|
274 |
+
"loss": 2.4103,
|
275 |
+
"step": 3800
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.35714285714285715,
|
279 |
+
"grad_norm": 0.24288786947727203,
|
280 |
+
"learning_rate": 8.918145392100745e-06,
|
281 |
+
"loss": 2.4149,
|
282 |
+
"step": 3900
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.3663003663003663,
|
286 |
+
"grad_norm": 0.25948914885520935,
|
287 |
+
"learning_rate": 9.147109330280481e-06,
|
288 |
+
"loss": 2.4236,
|
289 |
+
"step": 4000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.37545787545787546,
|
293 |
+
"grad_norm": 0.26415252685546875,
|
294 |
+
"learning_rate": 9.37607326846022e-06,
|
295 |
+
"loss": 2.4355,
|
296 |
+
"step": 4100
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.38461538461538464,
|
300 |
+
"grad_norm": 0.3095110058784485,
|
301 |
+
"learning_rate": 9.605037206639955e-06,
|
302 |
+
"loss": 2.4275,
|
303 |
+
"step": 4200
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.39377289377289376,
|
307 |
+
"grad_norm": 0.23992522060871124,
|
308 |
+
"learning_rate": 9.834001144819692e-06,
|
309 |
+
"loss": 2.4135,
|
310 |
+
"step": 4300
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.40293040293040294,
|
314 |
+
"grad_norm": 0.31741300225257874,
|
315 |
+
"learning_rate": 1.0062965082999428e-05,
|
316 |
+
"loss": 2.4148,
|
317 |
+
"step": 4400
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.41208791208791207,
|
321 |
+
"grad_norm": 0.26608923077583313,
|
322 |
+
"learning_rate": 1.0291929021179166e-05,
|
323 |
+
"loss": 2.4186,
|
324 |
+
"step": 4500
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.42124542124542125,
|
328 |
+
"grad_norm": 0.23381298780441284,
|
329 |
+
"learning_rate": 1.0520892959358902e-05,
|
330 |
+
"loss": 2.4124,
|
331 |
+
"step": 4600
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.43040293040293043,
|
335 |
+
"grad_norm": 0.42702221870422363,
|
336 |
+
"learning_rate": 1.0749856897538637e-05,
|
337 |
+
"loss": 2.412,
|
338 |
+
"step": 4700
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.43956043956043955,
|
342 |
+
"grad_norm": 0.24715246260166168,
|
343 |
+
"learning_rate": 1.0978820835718376e-05,
|
344 |
+
"loss": 2.4138,
|
345 |
+
"step": 4800
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.44871794871794873,
|
349 |
+
"grad_norm": 0.42784959077835083,
|
350 |
+
"learning_rate": 1.1207784773898111e-05,
|
351 |
+
"loss": 2.3953,
|
352 |
+
"step": 4900
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.45787545787545786,
|
356 |
+
"grad_norm": 0.25161468982696533,
|
357 |
+
"learning_rate": 1.143674871207785e-05,
|
358 |
+
"loss": 2.3898,
|
359 |
+
"step": 5000
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.46703296703296704,
|
363 |
+
"grad_norm": 0.2743668258190155,
|
364 |
+
"learning_rate": 1.1665712650257585e-05,
|
365 |
+
"loss": 2.4119,
|
366 |
+
"step": 5100
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.47619047619047616,
|
370 |
+
"grad_norm": 0.28708940744400024,
|
371 |
+
"learning_rate": 1.1894676588437321e-05,
|
372 |
+
"loss": 2.4039,
|
373 |
+
"step": 5200
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.48534798534798534,
|
377 |
+
"grad_norm": 0.31553077697753906,
|
378 |
+
"learning_rate": 1.212364052661706e-05,
|
379 |
+
"loss": 2.3987,
|
380 |
+
"step": 5300
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.4945054945054945,
|
384 |
+
"grad_norm": 0.2647671103477478,
|
385 |
+
"learning_rate": 1.2352604464796796e-05,
|
386 |
+
"loss": 2.4103,
|
387 |
+
"step": 5400
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.5036630036630036,
|
391 |
+
"grad_norm": 0.2579342722892761,
|
392 |
+
"learning_rate": 1.2581568402976532e-05,
|
393 |
+
"loss": 2.3984,
|
394 |
+
"step": 5500
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.5128205128205128,
|
398 |
+
"grad_norm": 0.365583211183548,
|
399 |
+
"learning_rate": 1.281053234115627e-05,
|
400 |
+
"loss": 2.3927,
|
401 |
+
"step": 5600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.521978021978022,
|
405 |
+
"grad_norm": 0.3024210035800934,
|
406 |
+
"learning_rate": 1.3039496279336006e-05,
|
407 |
+
"loss": 2.3956,
|
408 |
+
"step": 5700
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.5311355311355311,
|
412 |
+
"grad_norm": 0.31242358684539795,
|
413 |
+
"learning_rate": 1.3268460217515742e-05,
|
414 |
+
"loss": 2.3891,
|
415 |
+
"step": 5800
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.5402930402930403,
|
419 |
+
"grad_norm": 0.31390878558158875,
|
420 |
+
"learning_rate": 1.349742415569548e-05,
|
421 |
+
"loss": 2.3865,
|
422 |
+
"step": 5900
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.5494505494505495,
|
426 |
+
"grad_norm": 0.3613923192024231,
|
427 |
+
"learning_rate": 1.3726388093875217e-05,
|
428 |
+
"loss": 2.3919,
|
429 |
+
"step": 6000
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.5586080586080586,
|
433 |
+
"grad_norm": 0.30997100472450256,
|
434 |
+
"learning_rate": 1.3955352032054951e-05,
|
435 |
+
"loss": 2.3846,
|
436 |
+
"step": 6100
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.5677655677655677,
|
440 |
+
"grad_norm": 0.46809422969818115,
|
441 |
+
"learning_rate": 1.418431597023469e-05,
|
442 |
+
"loss": 2.3826,
|
443 |
+
"step": 6200
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.5769230769230769,
|
447 |
+
"grad_norm": 0.2764199674129486,
|
448 |
+
"learning_rate": 1.4413279908414425e-05,
|
449 |
+
"loss": 2.3893,
|
450 |
+
"step": 6300
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.5860805860805861,
|
454 |
+
"grad_norm": 0.3119123578071594,
|
455 |
+
"learning_rate": 1.4642243846594162e-05,
|
456 |
+
"loss": 2.3756,
|
457 |
+
"step": 6400
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.5952380952380952,
|
461 |
+
"grad_norm": 0.3249548375606537,
|
462 |
+
"learning_rate": 1.48712077847739e-05,
|
463 |
+
"loss": 2.3898,
|
464 |
+
"step": 6500
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.6043956043956044,
|
468 |
+
"grad_norm": 0.35118070244789124,
|
469 |
+
"learning_rate": 1.5100171722953636e-05,
|
470 |
+
"loss": 2.3845,
|
471 |
+
"step": 6600
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.6135531135531136,
|
475 |
+
"grad_norm": 0.307021826505661,
|
476 |
+
"learning_rate": 1.532913566113337e-05,
|
477 |
+
"loss": 2.3841,
|
478 |
+
"step": 6700
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.6227106227106227,
|
482 |
+
"grad_norm": 0.29582327604293823,
|
483 |
+
"learning_rate": 1.555809959931311e-05,
|
484 |
+
"loss": 2.3842,
|
485 |
+
"step": 6800
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.6318681318681318,
|
489 |
+
"grad_norm": 0.349324107170105,
|
490 |
+
"learning_rate": 1.5787063537492846e-05,
|
491 |
+
"loss": 2.3843,
|
492 |
+
"step": 6900
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.6410256410256411,
|
496 |
+
"grad_norm": 0.31297802925109863,
|
497 |
+
"learning_rate": 1.6016027475672583e-05,
|
498 |
+
"loss": 2.3819,
|
499 |
+
"step": 7000
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.6501831501831502,
|
503 |
+
"grad_norm": 0.27480989694595337,
|
504 |
+
"learning_rate": 1.624499141385232e-05,
|
505 |
+
"loss": 2.3553,
|
506 |
+
"step": 7100
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.6593406593406593,
|
510 |
+
"grad_norm": 0.33451569080352783,
|
511 |
+
"learning_rate": 1.6473955352032055e-05,
|
512 |
+
"loss": 2.3906,
|
513 |
+
"step": 7200
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.6684981684981685,
|
517 |
+
"grad_norm": 0.27567845582962036,
|
518 |
+
"learning_rate": 1.670291929021179e-05,
|
519 |
+
"loss": 2.3804,
|
520 |
+
"step": 7300
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.6776556776556777,
|
524 |
+
"grad_norm": 0.4053930938243866,
|
525 |
+
"learning_rate": 1.693188322839153e-05,
|
526 |
+
"loss": 2.376,
|
527 |
+
"step": 7400
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.6868131868131868,
|
531 |
+
"grad_norm": 0.30826249718666077,
|
532 |
+
"learning_rate": 1.7160847166571267e-05,
|
533 |
+
"loss": 2.3797,
|
534 |
+
"step": 7500
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.6959706959706959,
|
538 |
+
"grad_norm": 0.3094613254070282,
|
539 |
+
"learning_rate": 1.7389811104751004e-05,
|
540 |
+
"loss": 2.3781,
|
541 |
+
"step": 7600
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.7051282051282052,
|
545 |
+
"grad_norm": 0.2915087342262268,
|
546 |
+
"learning_rate": 1.761877504293074e-05,
|
547 |
+
"loss": 2.3694,
|
548 |
+
"step": 7700
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.7142857142857143,
|
552 |
+
"grad_norm": 0.276742547750473,
|
553 |
+
"learning_rate": 1.7847738981110476e-05,
|
554 |
+
"loss": 2.3586,
|
555 |
+
"step": 7800
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.7234432234432234,
|
559 |
+
"grad_norm": 0.2940647602081299,
|
560 |
+
"learning_rate": 1.8076702919290216e-05,
|
561 |
+
"loss": 2.377,
|
562 |
+
"step": 7900
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.7326007326007326,
|
566 |
+
"grad_norm": 0.3118553161621094,
|
567 |
+
"learning_rate": 1.8305666857469952e-05,
|
568 |
+
"loss": 2.363,
|
569 |
+
"step": 8000
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.7417582417582418,
|
573 |
+
"grad_norm": 0.29962947964668274,
|
574 |
+
"learning_rate": 1.8534630795649685e-05,
|
575 |
+
"loss": 2.3627,
|
576 |
+
"step": 8100
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.7509157509157509,
|
580 |
+
"grad_norm": 0.38140273094177246,
|
581 |
+
"learning_rate": 1.8763594733829424e-05,
|
582 |
+
"loss": 2.3629,
|
583 |
+
"step": 8200
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.76007326007326,
|
587 |
+
"grad_norm": 0.29137617349624634,
|
588 |
+
"learning_rate": 1.899255867200916e-05,
|
589 |
+
"loss": 2.3676,
|
590 |
+
"step": 8300
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.7692307692307693,
|
594 |
+
"grad_norm": 0.3427438735961914,
|
595 |
+
"learning_rate": 1.9221522610188897e-05,
|
596 |
+
"loss": 2.3621,
|
597 |
+
"step": 8400
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.7783882783882784,
|
601 |
+
"grad_norm": 0.30704206228256226,
|
602 |
+
"learning_rate": 1.9450486548368633e-05,
|
603 |
+
"loss": 2.3508,
|
604 |
+
"step": 8500
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.7875457875457875,
|
608 |
+
"grad_norm": 0.34004345536231995,
|
609 |
+
"learning_rate": 1.967945048654837e-05,
|
610 |
+
"loss": 2.3499,
|
611 |
+
"step": 8600
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.7967032967032966,
|
615 |
+
"grad_norm": 0.39838922023773193,
|
616 |
+
"learning_rate": 1.9908414424728106e-05,
|
617 |
+
"loss": 2.3544,
|
618 |
+
"step": 8700
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.8058608058608059,
|
622 |
+
"grad_norm": 0.3006947636604309,
|
623 |
+
"learning_rate": 1.9962812240794344e-05,
|
624 |
+
"loss": 2.36,
|
625 |
+
"step": 8800
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.815018315018315,
|
629 |
+
"grad_norm": 0.3303048610687256,
|
630 |
+
"learning_rate": 1.973655456746745e-05,
|
631 |
+
"loss": 2.3457,
|
632 |
+
"step": 8900
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.8241758241758241,
|
636 |
+
"grad_norm": 0.31546804308891296,
|
637 |
+
"learning_rate": 1.9309362692059617e-05,
|
638 |
+
"loss": 2.3522,
|
639 |
+
"step": 9000
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.8333333333333334,
|
643 |
+
"grad_norm": 0.3123638927936554,
|
644 |
+
"learning_rate": 1.869005261390877e-05,
|
645 |
+
"loss": 2.3552,
|
646 |
+
"step": 9100
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.8424908424908425,
|
650 |
+
"grad_norm": 0.36076679825782776,
|
651 |
+
"learning_rate": 1.789140509396394e-05,
|
652 |
+
"loss": 2.3434,
|
653 |
+
"step": 9200
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.8516483516483516,
|
657 |
+
"grad_norm": 0.3313111960887909,
|
658 |
+
"learning_rate": 1.69299018970067e-05,
|
659 |
+
"loss": 2.3782,
|
660 |
+
"step": 9300
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.8608058608058609,
|
664 |
+
"grad_norm": 0.3116619884967804,
|
665 |
+
"learning_rate": 1.582538565590479e-05,
|
666 |
+
"loss": 2.3398,
|
667 |
+
"step": 9400
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.86996336996337,
|
671 |
+
"grad_norm": 0.33244970440864563,
|
672 |
+
"learning_rate": 1.4600650377311523e-05,
|
673 |
+
"loss": 2.3443,
|
674 |
+
"step": 9500
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.8791208791208791,
|
678 |
+
"grad_norm": 0.3026200830936432,
|
679 |
+
"learning_rate": 1.3280971039583906e-05,
|
680 |
+
"loss": 2.3442,
|
681 |
+
"step": 9600
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.8882783882783882,
|
685 |
+
"grad_norm": 0.35270655155181885,
|
686 |
+
"learning_rate": 1.1893581990651848e-05,
|
687 |
+
"loss": 2.337,
|
688 |
+
"step": 9700
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.8974358974358975,
|
692 |
+
"grad_norm": 0.34508877992630005,
|
693 |
+
"learning_rate": 1.046711491019129e-05,
|
694 |
+
"loss": 2.3637,
|
695 |
+
"step": 9800
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.9065934065934066,
|
699 |
+
"grad_norm": 0.3241460919380188,
|
700 |
+
"learning_rate": 9.031007934929237e-06,
|
701 |
+
"loss": 2.3529,
|
702 |
+
"step": 9900
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.9157509157509157,
|
706 |
+
"grad_norm": 0.3581334054470062,
|
707 |
+
"learning_rate": 7.6148981410179966e-06,
|
708 |
+
"loss": 2.3524,
|
709 |
+
"step": 10000
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.924908424908425,
|
713 |
+
"grad_norm": 0.34301674365997314,
|
714 |
+
"learning_rate": 6.248009920877591e-06,
|
715 |
+
"loss": 2.3532,
|
716 |
+
"step": 10100
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.9340659340659341,
|
720 |
+
"grad_norm": 0.30783897638320923,
|
721 |
+
"learning_rate": 4.95855187663155e-06,
|
722 |
+
"loss": 2.3507,
|
723 |
+
"step": 10200
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.9432234432234432,
|
727 |
+
"grad_norm": 0.40239864587783813,
|
728 |
+
"learning_rate": 3.773134676503629e-06,
|
729 |
+
"loss": 2.3488,
|
730 |
+
"step": 10300
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.9523809523809523,
|
734 |
+
"grad_norm": 0.3629162907600403,
|
735 |
+
"learning_rate": 2.7162218879283174e-06,
|
736 |
+
"loss": 2.3507,
|
737 |
+
"step": 10400
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.9615384615384616,
|
741 |
+
"grad_norm": 0.3233180046081543,
|
742 |
+
"learning_rate": 1.8096251205843685e-06,
|
743 |
+
"loss": 2.3369,
|
744 |
+
"step": 10500
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.9706959706959707,
|
748 |
+
"grad_norm": 0.3085317313671112,
|
749 |
+
"learning_rate": 1.0720538981326557e-06,
|
750 |
+
"loss": 2.3495,
|
751 |
+
"step": 10600
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.9798534798534798,
|
755 |
+
"grad_norm": 0.332645446062088,
|
756 |
+
"learning_rate": 5.187295479980136e-07,
|
757 |
+
"loss": 2.3448,
|
758 |
+
"step": 10700
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.989010989010989,
|
762 |
+
"grad_norm": 0.3299219012260437,
|
763 |
+
"learning_rate": 1.6107107738836835e-07,
|
764 |
+
"loss": 2.3427,
|
765 |
+
"step": 10800
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.9981684981684982,
|
769 |
+
"grad_norm": 0.41852566599845886,
|
770 |
+
"learning_rate": 6.4595181575910496e-09,
|
771 |
+
"loss": 2.3397,
|
772 |
+
"step": 10900
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 1.0,
|
776 |
+
"step": 10920,
|
777 |
+
"total_flos": 1.59124040841796e+18,
|
778 |
+
"train_loss": 2.428147726093893,
|
779 |
+
"train_runtime": 2109.7476,
|
780 |
+
"train_samples_per_second": 82.811,
|
781 |
+
"train_steps_per_second": 5.176
|
782 |
+
}
|
783 |
+
],
|
784 |
+
"logging_steps": 100,
|
785 |
+
"max_steps": 10920,
|
786 |
+
"num_input_tokens_seen": 0,
|
787 |
+
"num_train_epochs": 1,
|
788 |
+
"save_steps": 500,
|
789 |
+
"stateful_callbacks": {
|
790 |
+
"TrainerControl": {
|
791 |
+
"args": {
|
792 |
+
"should_epoch_stop": false,
|
793 |
+
"should_evaluate": false,
|
794 |
+
"should_log": false,
|
795 |
+
"should_save": true,
|
796 |
+
"should_training_stop": true
|
797 |
+
},
|
798 |
+
"attributes": {}
|
799 |
+
}
|
800 |
+
},
|
801 |
+
"total_flos": 1.59124040841796e+18,
|
802 |
+
"train_batch_size": 2,
|
803 |
+
"trial_name": null,
|
804 |
+
"trial_params": null
|
805 |
+
}
|