gpt2moe_dom_het_tok_1000mb
GPT2MoE (custom) trained on a set of languages where English dominates the model and tokenizer training data sets English is dominant in the model training set (3000mb of English and 500mb (both with byte premium scaling) for all other languages) English is dominant in the tokenizer training set (300mb of English and 50mb (both with byte premium scaling) for all other languages)
Training results
Based on 3 W&B runs named gpt2moe_dom_het_tok_1000mb in sjzahra-ri-se/huggingface.
Loss: 2 run(s); warmup: 2; GPU telemetry: 3 run(s).
Training hardware: Trained on a single H100 GPU.
Languages
Greek, Turkish, Estonian, English, Arabic
Training metadata
| Parameter | Value |
|---|---|
| Runs included | 3 |
| Training hardware | Trained on a single H100 GPU |
| Total epochs | 9.99977 |
| Total training steps | 360,651 |
| Total training FLOPs | 1.2063e+19 |
| Learning rate | 0.0001 |
| Train batch size | 8 |
| Eval batch size | 4 |
| Gradient accumulation | 8 |
| Weight decay | 0.01 |
| Warmup steps | 36065 |
| Warmup ratio | 0 |
| Optimizer | adamw_torch |
| Scheduler | linear |
| Max sequence length | 20 |
| Precision | False |
| Seed | 42 |
| GPU utilization | mean 37.1%, peak 50.0% |
W&B runs: 0kj19n5x, lqv5nimi, nmuu5s3r
Generated 2026-11-17T17:17:59+00:00.
- Downloads last month
- 10
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐ Ask for provider support
Collection including NLP-RISE/gpt2moe_dom_het_tok_1000mb
Collection
Trained on est_latn, arb_arab, ell_grek, tur_latn, and eng_latn (heterogeneous set 1). โข 6 items โข Updated
