Topic Classification on SIB200 EN
78.9AccuracyLlama-3.2-3B
Evaluation results
| Method | Links | |
|---|---|---|
| Llama-3.2-3BTraining Status=Continued pretraining, Tokenizer Strategy=prune+ext (16k, naive)2025.12 | 78.9 | |
| Llama-3.2-3BTraining Status=Continued pretraining, Tokenizer Strategy=prune+ext (16k, ours)2025.12 | 77.9 | |
| Llama-3.2-1BTraining Status=No training2025.12 | 77 | |
| Llama-3.2-3BTraining Status=Continued pretraining, Tokenizer Strategy=default2025.12 | 77 | |
| Llama-3.2-3BTraining Status=No training2025.12 | 76 | |
| Llama-3.2-1BTraining Status=Continued pretraining, Tokenizer Strategy=default2025.12 | 74.5 | |
| Llama-3.2-1BTraining Status=Continued pretraining, Tokenizer Strategy=prune+ext (16k, naive)2025.12 | 73 | |
| Llama-3.2-1BTraining Status=Continued pretraining, Tokenizer Strategy=prune+ext (16k, ours)2025.12 | 73 |