Update README.md
Browse files
README.md
CHANGED
@@ -20,3 +20,74 @@ tags:
|
|
20 |
This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
|
21 |
|
22 |
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
20 |
This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
|
21 |
|
22 |
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
23 |
+
|
24 |
+
Step Training Loss Validation Loss
|
25 |
+
10 1.969100 2.004664
|
26 |
+
20 1.912000 2.000807
|
27 |
+
30 1.973500 1.997514
|
28 |
+
40 1.987000 1.995450
|
29 |
+
50 2.035200 1.992437
|
30 |
+
60 1.947100 1.989926
|
31 |
+
70 1.913200 1.988254
|
32 |
+
80 1.984200 1.985698
|
33 |
+
90 1.829400 1.984344
|
34 |
+
100 1.925600 1.982281
|
35 |
+
110 1.924200 1.980534
|
36 |
+
120 1.946400 1.979197
|
37 |
+
130 1.886500 1.977808
|
38 |
+
140 1.911200 1.976381
|
39 |
+
150 1.855700 1.974918
|
40 |
+
160 1.906900 1.973701
|
41 |
+
170 1.827500 1.972471
|
42 |
+
180 1.905400 1.972400
|
43 |
+
190 1.864500 1.972158
|
44 |
+
200 1.974000 1.971486
|
45 |
+
210 2.020100 1.970601
|
46 |
+
220 1.835600 1.969159
|
47 |
+
230 1.873000 1.969961
|
48 |
+
240 1.853200 1.968564
|
49 |
+
250 1.892800 1.968765
|
50 |
+
260 1.808400 1.967971
|
51 |
+
270 1.818600 1.967605
|
52 |
+
280 1.866600 1.967552
|
53 |
+
290 1.761000 1.966953
|
54 |
+
300 1.860300 1.966536
|
55 |
+
310 1.793400 1.966086
|
56 |
+
320 1.814500 1.965425
|
57 |
+
330 1.978200 1.965850
|
58 |
+
340 1.868600 1.965540
|
59 |
+
350 1.834300 1.966008
|
60 |
+
360 1.822400 1.966800
|
61 |
+
370 1.896100 1.968465
|
62 |
+
380 1.883600 1.967751
|
63 |
+
390 1.810500 1.967558
|
64 |
+
400 1.808000 1.967848
|
65 |
+
410 1.771100 1.968701
|
66 |
+
420 1.877800 1.967933
|
67 |
+
430 1.838300 1.968531
|
68 |
+
440 1.717500 1.968299
|
69 |
+
450 1.848500 1.969323
|
70 |
+
460 1.794400 1.969219
|
71 |
+
470 1.864300 1.969595
|
72 |
+
480 1.768400 1.968718
|
73 |
+
490 1.682400 1.969312
|
74 |
+
500 1.835200 1.967268
|
75 |
+
510 1.754400 1.968593
|
76 |
+
520 1.870700 1.968871
|
77 |
+
530 1.810000 1.972527
|
78 |
+
540 1.813400 1.972523
|
79 |
+
550 1.767600 1.973855
|
80 |
+
560 1.874200 1.974136
|
81 |
+
570 1.791200 1.973645
|
82 |
+
580 1.904100 1.974470
|
83 |
+
590 1.792400 1.972956
|
84 |
+
600 1.841000 1.974010
|
85 |
+
610 1.769700 1.974349
|
86 |
+
620 2.016400 1.974549
|
87 |
+
630 1.804100 1.974385
|
88 |
+
640 1.891100 1.975045
|
89 |
+
650 1.785100 1.975142
|
90 |
+
660 1.760900 1.974409
|
91 |
+
670 1.822400 1.974042
|
92 |
+
680 1.783800 1.974177
|
93 |
+
690 1.757100 1.974227
|