language: - en tags: - pretraining - causal-lm - nanochat license: mit datasets: - karpathy/fineweb-edu-100b-shuffle - HuggingFaceTB/smol-smoltalk - openai/gsm8k - allenai/ai2_arc - cais/mmlu metrics: - accuracy
Tag: d20 Step: 21400
Exported from nanochat.