Hugging Face's logo

GoudaAI
/
Cheddar-2.0

Text Generation
Safetensors
English
gouda-cheddar
custom_code
Model card Files
xet
 Community
Erbium08 commited on
Commit
c50e897
·
verified ·
1 Parent(s): a1218a6

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -6
README.md CHANGED
@@ -79,12 +79,12 @@ Across evaluation benchmarks, the gouda both pretrained and instruct models disp
79
  ### Cheddar-2.0 (302m, 5.6B tokens) vs Pythia-410m (300B tokens) - Closest Pythia Model, raw_acc
80
  | Task | Gouda-2.0 | Pythia-160m | Gap |
81
  |---------------|-----------|-------------|------|
82
- | ARC-Easy | .518 | .521 | +7.6 |
83
- | ARC-Challenge | .222 | .213 | +3.8 |
84
- | SciQ | .762 | .811 | +1.3 |
85
- | PIQA | .631 | .668 | +0.2 |
86
- | WinoGrande | .515 | .537 | -1.9 |
87
- | LAMBADA | .262 | .516 | -6.2 |
88
 
89
 
90
  ### PRETRAINED MODELS - Name (Parameters / Tokens)
 
79
  ### Cheddar-2.0 (302m, 5.6B tokens) vs Pythia-410m (300B tokens) - Closest Pythia Model, raw_acc
80
  | Task | Gouda-2.0 | Pythia-160m | Gap |
81
  |---------------|-----------|-------------|------|
82
+ | ARC-Easy | .518 | .521 | -0.3 |
83
+ | ARC-Challenge | .222 | .213 | +0.9 |
84
+ | SciQ | .762 | .811 | -4.9 |
85
+ | PIQA | .631 | .668 | -3.7 |
86
+ | WinoGrande | .515 | .537 | -2.2 |
87
+ | LAMBADA | .262 | .516 | -25.4 |
88
 
89
 
90
  ### PRETRAINED MODELS - Name (Parameters / Tokens)