Perplexity (ctx 8192)
BF16:
I1005 10:12:54.337877 967332 eval_ppl.py:66] wikitext2 perplexity: 6.49992036819458
I1005 10:17:19.718732 967332 eval_ppl.py:66] c4 perplexity: 8.022844314575195
2 Bit QTIP (this model):
I1005 10:13:20.523440 967333 eval_ppl.py:66] wikitext2 perplexity: 7.8186869621276855
I1005 10:19:34.002755 967333 eval_ppl.py:66] c4 perplexity: 9.199813842773438
Zeroshot Results with lm_eval 0.3.0
BF16:
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.5196|± |0.0146|
| | |acc_norm|0.5512|± |0.0145|
|arc_easy | 0|acc |0.8182|± |0.0079|
| | |acc_norm|0.7980|± |0.0082|
|boolq | 1|acc |0.8410|± |0.0064|
|piqa | 0|acc |0.8003|± |0.0093|
| | |acc_norm|0.8090|± |0.0092|
|winogrande | 0|acc |0.7380|± |0.0124|
2 Bit QTIP (this model):
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.4676|± |0.0146|
| | |acc_norm|0.5077|± |0.0146|
|arc_easy | 0|acc |0.7769|± |0.0085|
| | |acc_norm|0.7551|± |0.0088|
|boolq | 1|acc |0.8101|± |0.0069|
|piqa | 0|acc |0.7775|± |0.0097|
| | |acc_norm|0.7829|± |0.0096|
|winogrande | 0|acc |0.6985|± |0.0129|