Machine Translation on FLORES ET-EN v1 (test)
0.831COMETLlama-3.2-3B
Evaluation results
| Method | Links | |
|---|---|---|
| Llama-3.2-3BTraining Status=Continued pretraining, Tokenizer Strategy=prune+ext (16k, naive), Training Tokens=5.4B2025.12 | 0.831 | |
| Llama-3.2-3BTraining Status=Continued pretraining, Tokenizer Strategy=default, Training Tokens=7.2B2025.12 | 0.83 | |
| Llama-3.2-3BTraining Status=Continued pretraining, Tokenizer Strategy=prune+ext (16k, ours), Training Tokens=5.3B2025.12 | 0.829 | |
| Llama-3.2-1BTraining Status=Continued pretraining, Tokenizer Strategy=prune+ext (16k, naive), Training Tokens=5.4B2025.12 | 0.809 | |
| Llama-3.2-1BTraining Status=Continued pretraining, Tokenizer Strategy=default, Training Tokens=7.2B2025.12 | 0.803 | |
| Llama-3.2-1BTraining Status=Continued pretraining, Tokenizer Strategy=prune+ext (16k, ours), Training Tokens=5.3B2025.12 | 0.803 | |
| Llama-3.2-3BTraining Status=No training, Tokenizer Strategy=default, Training Tokens=02025.12 | 0.796 | |
| Llama-3.2-1BTraining Status=No training, Tokenizer Strategy=default, Training Tokens=02025.12 | 0.701 |