MaxLSB commited on
Commit
774ee5f
·
verified ·
1 Parent(s): 44de1a7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -14
README.md CHANGED
@@ -36,25 +36,25 @@ We used LightEval for evaluation, with custom tasks for the French benchmarks. T
36
 
37
  ### French Benchmark Scores
38
 
39
- | Benchmark | LFM2-1.2B | Luth-LFM2-1.2B |
40
  |-------------------|------------------|-----------------------|
41
- | ifeval-fr (strict prompt) | 53.60 | <u>60.44</u> |
42
- | gpqa-fr | 25.77 | <u>27.02</u> |
43
- | mmlu-fr | 47.59 | <u>47.98</u> |
44
- | math-500-fr | 35.80 | <u>47.20</u> |
45
- | arc-chall-fr | <u>39.44</u> | 39.01 |
46
- | hellaswag-fr | 33.05 | <u>36.76</u> |
47
 
48
  ### English Benchmark Scores
49
 
50
- | Benchmark | LFM2-1.2B | Luth-LFM2-1.2B |
51
  |-------------------|------------------|-----------------------|
52
- | ifeval-en (strict prompt)| 70.43 | <u>70.61</u> |
53
- | gpqa-en | 26.68 | <u>28.21</u> |
54
- | mmlu-en | <u>55.18</u> | 54.59 |
55
- | math-500-en | 44.60 | <u>50.20</u> |
56
- | arc-chall-en | 43.09 | <u>43.26</u> |
57
- | hellaswag-en | 57.64 | <u>58.46</u> |
58
 
59
  ## Code Example
60
 
 
36
 
37
  ### French Benchmark Scores
38
 
39
+ | **Benchmark** | **LFM2-1.2B** | **Luth-LFM2-1.2B** |
40
  |-------------------|------------------|-----------------------|
41
+ | **IFEval-fr (strict prompt)** | 53.60 | <u>60.44</u> |
42
+ | **GPQA-fr** | 25.77 | <u>27.02</u> |
43
+ | **MMLU-fr** | 47.59 | <u>47.98</u> |
44
+ | **MATH-500-fr** | 35.80 | <u>47.20</u> |
45
+ | **Arc-Chall-fr** | <u>39.44</u> | 39.01 |
46
+ | **Hellaswag-fr** | 33.05 | <u>36.76</u> |
47
 
48
  ### English Benchmark Scores
49
 
50
+ | **Benchmark** | **LFM2-1.2B** | **Luth-LFM2-1.2B** |
51
  |-------------------|------------------|-----------------------|
52
+ | **IFEval-en (strict prompt)**| 70.43 | <u>70.61</u> |
53
+ | **GPQA-en** | 26.68 | <u>28.21</u> |
54
+ | **MMLU-en** | <u>55.18</u> | 54.59 |
55
+ | **MATH-500-en** | 44.60 | <u>50.20</u> |
56
+ | **Arc-Chall-en** | 43.09 | <u>43.26</u> |
57
+ | **Hellaswag-en** | 57.64 | <u>58.46</u> |
58
 
59
  ## Code Example
60