Update README.md
Browse files
README.md
CHANGED
|
@@ -21,17 +21,27 @@ It achieves the following results on the evaluation set:
|
|
| 21 |
- Accuracy: 0.4514
|
| 22 |
- Num Input Tokens Seen: 1975517184
|
| 23 |
|
| 24 |
-
## Model description
|
| 25 |
|
| 26 |
-
|
| 27 |
|
| 28 |
-
|
| 29 |
|
| 30 |
-
More information needed
|
| 31 |
|
| 32 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
|
| 34 |
-
More information needed
|
| 35 |
|
| 36 |
## Training procedure
|
| 37 |
|
|
|
|
| 21 |
- Accuracy: 0.4514
|
| 22 |
- Num Input Tokens Seen: 1975517184
|
| 23 |
|
|
|
|
| 24 |
|
| 25 |
+
## Quick eval
|
| 26 |
|
| 27 |
+
Quick eval for: pszemraj/jamba-H1024_L12-v0.13-KIx2
|
| 28 |
|
|
|
|
| 29 |
|
| 30 |
+
bootstrapping for stddev: perplexity
|
| 31 |
+
hf (pretrained=pszemraj/jamba-H1024_L12-v0.13-KIx2,trust_remote_code=True,dtype=float), gen_kwargs: (None), limit: 0.9999, num_fewshot: None, batch_size: 8
|
| 32 |
+
| Tasks |Version|Filter|n-shot| Metric | Value | |Stderr|
|
| 33 |
+
|--------------|------:|------|-----:|----------|-------:|---|-----:|
|
| 34 |
+
|winogrande | 1|none | 0|acc | 0.5067|± |0.0141|
|
| 35 |
+
|piqa | 1|none | 0|acc | 0.5912|± |0.0138|
|
| 36 |
+
| | |none | 0|acc_norm | 0.5951|± |0.0138|
|
| 37 |
+
|openbookqa | 1|none | 0|acc | 0.1800|± |0.0172|
|
| 38 |
+
| | |none | 0|acc_norm | 0.2920|± |0.0204|
|
| 39 |
+
|lambada_openai| 1|none | 0|perplexity|103.1241|± |8.5843|
|
| 40 |
+
| | |none | 0|acc | 0.2502|± |0.0122|
|
| 41 |
+
|boolq | 2|none | 0|acc | 0.6196|± |0.0136|
|
| 42 |
+
|arc_easy | 1|none | 0|acc | 0.3836|± |0.0137|
|
| 43 |
+
| | |none | 0|acc_norm | 0.3694|± |0.0136|
|
| 44 |
|
|
|
|
| 45 |
|
| 46 |
## Training procedure
|
| 47 |
|