zechen-nlp commited on
Commit
b4b0d88
Β·
verified Β·
1 Parent(s): 6ef3995

Update Automated MNLP evaluation report (2026-06-10)

Browse files
Files changed (1) hide show
  1. EVAL_REPORT.md +11 -13
EVAL_REPORT.md CHANGED
@@ -2,23 +2,23 @@
2
 
3
  - **Model repo:** [`cs-552-2026-MandMP/general_knowledge_model`](https://huggingface.co/cs-552-2026-MandMP/general_knowledge_model)
4
  - **Owner(s):** group **MandMP**
5
- - **Generated at:** 2026-06-09T09:44:07+00:00 (UTC)
6
  - **Pipeline:** [mnlp-project-ci](https://github.com/eric11eca/mnlp-project-ci)
7
 
8
  _This PR is opened automatically by the course CI. It is **non-blocking** β€” you do not need to merge it. The next nightly run will refresh this file._
9
 
10
  ## Evaluated checkpoint
11
 
12
- - **Commit:** [`b4ecf2d`](https://huggingface.co/cs-552-2026-MandMP/general_knowledge_model/commit/b4ecf2d312082d6e0c02c5de176f1ae2d120e644)
13
- - **Message:** Set temperature=0 for deterministic inference
14
- - **Committed:** 2026-06-08T11:33:01+00:00
15
 
16
  ## Summary
17
 
18
  | Benchmark | Accuracy | Status |
19
  |---|---:|---|
20
  | Math | β€” | not run |
21
- | Knowledge | 0.3100 | ok |
22
  | Multilingual | β€” | not run |
23
  | Safety | β€” | not run |
24
 
@@ -30,28 +30,26 @@ _Prompts are intentionally omitted to avoid revealing benchmark contents. For mu
30
 
31
  **Correct** (1 shown)
32
 
33
- - **reference**: `B`
34
  - **overall** (1/1 completions correct)
35
- - **extracted** (βœ“): `B`
36
  - **completion**:
37
 
38
  ```text
39
  <think>
40
 
41
- \boxed{B}
42
  ```
43
 
44
  **Incorrect** (1 shown)
45
 
46
- - **reference**: `A`
47
  - **overall** (0/1 completions correct)
48
- - **extracted** (βœ—): `C`
49
  - **completion**:
50
 
51
  ```text
52
  <think>
53
 
54
- </think>
55
-
56
- \boxed{C}
57
  ```
 
2
 
3
  - **Model repo:** [`cs-552-2026-MandMP/general_knowledge_model`](https://huggingface.co/cs-552-2026-MandMP/general_knowledge_model)
4
  - **Owner(s):** group **MandMP**
5
+ - **Generated at:** 2026-06-10T21:20:55+00:00 (UTC)
6
  - **Pipeline:** [mnlp-project-ci](https://github.com/eric11eca/mnlp-project-ci)
7
 
8
  _This PR is opened automatically by the course CI. It is **non-blocking** β€” you do not need to merge it. The next nightly run will refresh this file._
9
 
10
  ## Evaluated checkpoint
11
 
12
+ - **Commit:** [`1084b06`](https://huggingface.co/cs-552-2026-MandMP/general_knowledge_model/commit/1084b06be7e3d0b96b17e3abc8b0983713cf7ee4)
13
+ - **Message:** Fix temperature for Qwen3 thinking mode (0.6/top_p=0.95)
14
+ - **Committed:** 2026-06-09T15:20:11+00:00
15
 
16
  ## Summary
17
 
18
  | Benchmark | Accuracy | Status |
19
  |---|---:|---|
20
  | Math | β€” | not run |
21
+ | Knowledge | 0.3200 | ok |
22
  | Multilingual | β€” | not run |
23
  | Safety | β€” | not run |
24
 
 
30
 
31
  **Correct** (1 shown)
32
 
33
+ - **reference**: `C`
34
  - **overall** (1/1 completions correct)
35
+ - **extracted** (βœ“): `C`
36
  - **completion**:
37
 
38
  ```text
39
  <think>
40
 
41
+ \boxed{C}
42
  ```
43
 
44
  **Incorrect** (1 shown)
45
 
46
+ - **reference**: `C`
47
  - **overall** (0/1 completions correct)
48
+ - **extracted** (βœ—): `D`
49
  - **completion**:
50
 
51
  ```text
52
  <think>
53
 
54
+ \boxed{D}
 
 
55
  ```