Text Generation
Transformers
Safetensors
English
Italian
gpt2
1gpu-llm
single-gpu
trained-from-scratch
gpt2preln
bilingual
english
italian
pretraining
base-model
causal-lm
llm-nanochat
preln
decay-only
text-generation-inference
Instructions to use nazdef/1gpu-llm-small-en-it-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use nazdef/1gpu-llm-small-en-it-base with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="nazdef/1gpu-llm-small-en-it-base")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("nazdef/1gpu-llm-small-en-it-base") model = AutoModelForCausalLM.from_pretrained("nazdef/1gpu-llm-small-en-it-base") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use nazdef/1gpu-llm-small-en-it-base with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "nazdef/1gpu-llm-small-en-it-base" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "nazdef/1gpu-llm-small-en-it-base", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/nazdef/1gpu-llm-small-en-it-base
- SGLang
How to use nazdef/1gpu-llm-small-en-it-base with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "nazdef/1gpu-llm-small-en-it-base" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "nazdef/1gpu-llm-small-en-it-base", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "nazdef/1gpu-llm-small-en-it-base" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "nazdef/1gpu-llm-small-en-it-base", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use nazdef/1gpu-llm-small-en-it-base with Docker Model Runner:
docker model run hf.co/nazdef/1gpu-llm-small-en-it-base
| { | |
| "contains_mixed_model_types": false, | |
| "metric_recommendations": { | |
| "cloze_en_contains": { | |
| "checkpoint_name": "step_8600", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8600.pt", | |
| "direction": "max", | |
| "value": 0.14 | |
| }, | |
| "cloze_en_exact": { | |
| "checkpoint_name": "step_8050", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8050.pt", | |
| "direction": "max", | |
| "value": 0.0 | |
| }, | |
| "cloze_it_contains": { | |
| "checkpoint_name": "step_8350", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8350.pt", | |
| "direction": "max", | |
| "value": 0.26 | |
| }, | |
| "cloze_it_exact": { | |
| "checkpoint_name": "step_8050", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8050.pt", | |
| "direction": "max", | |
| "value": 0.0 | |
| }, | |
| "distinct_1": { | |
| "checkpoint_name": "step_8300", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8300.pt", | |
| "direction": "max", | |
| "value": 0.2651821862348178 | |
| }, | |
| "distinct_2": { | |
| "checkpoint_name": "step_8300", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8300.pt", | |
| "direction": "max", | |
| "value": 0.5284810126582279 | |
| }, | |
| "language_consistency_en": { | |
| "checkpoint_name": "step_8050", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8050.pt", | |
| "direction": "max", | |
| "value": 1.0 | |
| }, | |
| "language_consistency_it": { | |
| "checkpoint_name": "step_8150", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8150.pt", | |
| "direction": "max", | |
| "value": 0.875 | |
| }, | |
| "language_switch_rate_en": { | |
| "checkpoint_name": "step_8050", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8050.pt", | |
| "direction": "min", | |
| "value": 0.0 | |
| }, | |
| "language_switch_rate_it": { | |
| "checkpoint_name": "step_8050", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8050.pt", | |
| "direction": "min", | |
| "value": 0.0 | |
| }, | |
| "loop_rate": { | |
| "checkpoint_name": "step_8450", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8450.pt", | |
| "direction": "min", | |
| "value": 0.35 | |
| }, | |
| "ppl_en": { | |
| "checkpoint_name": "step_8800", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8800.pt", | |
| "direction": "min", | |
| "value": 115.77590188514857 | |
| }, | |
| "ppl_it": { | |
| "checkpoint_name": "step_8550", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8550.pt", | |
| "direction": "min", | |
| "value": 37.17054792048566 | |
| }, | |
| "ppl_mixed": { | |
| "checkpoint_name": "step_8600", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8600.pt", | |
| "direction": "min", | |
| "value": 121.06936368453971 | |
| }, | |
| "repeated_4gram_rate": { | |
| "checkpoint_name": "step_8300", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8300.pt", | |
| "direction": "min", | |
| "value": 0.75 | |
| }, | |
| "source_loss_books_en": { | |
| "checkpoint_name": "step_8400", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8400.pt", | |
| "direction": "min", | |
| "value": 4.808423723493304 | |
| }, | |
| "source_loss_books_it": { | |
| "checkpoint_name": "step_8600", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8600.pt", | |
| "direction": "min", | |
| "value": 4.5494671776181175 | |
| }, | |
| "source_loss_code": { | |
| "checkpoint_name": "step_8450", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8450.pt", | |
| "direction": "min", | |
| "value": 7.7475123087565105 | |
| }, | |
| "source_loss_web_en": { | |
| "checkpoint_name": "step_8650", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8650.pt", | |
| "direction": "min", | |
| "value": 5.791411550421464 | |
| }, | |
| "source_loss_web_it": { | |
| "checkpoint_name": "step_8650", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8650.pt", | |
| "direction": "min", | |
| "value": 5.6818381861636515 | |
| }, | |
| "source_loss_wiki_en": { | |
| "checkpoint_name": "step_8700", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8700.pt", | |
| "direction": "min", | |
| "value": 3.4160707647150215 | |
| }, | |
| "source_loss_wiki_it": { | |
| "checkpoint_name": "step_8550", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8550.pt", | |
| "direction": "min", | |
| "value": 3.4177459716796874 | |
| }, | |
| "val_loss_en": { | |
| "checkpoint_name": "step_8800", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8800.pt", | |
| "direction": "min", | |
| "value": 4.751656442318323 | |
| }, | |
| "val_loss_it": { | |
| "checkpoint_name": "step_8550", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8550.pt", | |
| "direction": "min", | |
| "value": 3.615516725133677 | |
| }, | |
| "val_loss_mixed": { | |
| "checkpoint_name": "step_8600", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8600.pt", | |
| "direction": "min", | |
| "value": 4.796363634940906 | |
| } | |
| }, | |
| "recommended_checkpoint": { | |
| "checkpoint_name": "step_8600", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8600.pt", | |
| "direction": "min", | |
| "value": 4.796363634940906 | |
| }, | |
| "recommended_metric": "val_loss_mixed", | |
| "rows": [ | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8050", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8050.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8050, | |
| "cloze_en_contains": 0.04, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.16, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.25430210325047803, | |
| "distinct_2": 0.49204771371769385, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 1.0, | |
| "language_consistency_it": 0.8, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.475, | |
| "model_type": "pretrained", | |
| "ppl_en": 160.98727346833914, | |
| "ppl_it": 49.3262033438898, | |
| "ppl_mixed": 148.87818824189094, | |
| "repeated_4gram_rate": 0.8, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 5.168394906180246, | |
| "source_loss_books_it": 4.828146798270089, | |
| "source_loss_code": 8.584260050455729, | |
| "source_loss_web_en": 6.076142963610198, | |
| "source_loss_web_it": 5.855331019351357, | |
| "source_loss_wiki_en": 3.9519438310102983, | |
| "source_loss_wiki_it": 3.720078430175781, | |
| "val_loss_en": 5.081325315079599, | |
| "val_loss_it": 3.8984554478379545, | |
| "val_loss_mixed": 5.003128443008814, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8100", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8100.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8100, | |
| "cloze_en_contains": 0.02, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.1, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.24681684622918706, | |
| "distinct_2": 0.47706422018348627, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.975, | |
| "language_consistency_it": 0.775, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.5, | |
| "model_type": "pretrained", | |
| "ppl_en": 133.99624078278714, | |
| "ppl_it": 37.37222258583596, | |
| "ppl_mixed": 146.63460264455082, | |
| "repeated_4gram_rate": 0.875, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 5.14651852562314, | |
| "source_loss_books_it": 4.574619838169643, | |
| "source_loss_code": 8.119071960449219, | |
| "source_loss_web_en": 5.854736328125, | |
| "source_loss_web_it": 5.742109599866365, | |
| "source_loss_wiki_en": 3.55948188088157, | |
| "source_loss_wiki_it": 3.5610482788085935, | |
| "val_loss_en": 4.897811745697597, | |
| "val_loss_it": 3.620927716864914, | |
| "val_loss_mixed": 4.987943796011118, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8150", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8150.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8150, | |
| "cloze_en_contains": 0.0, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.12, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.21788772597526165, | |
| "distinct_2": 0.4065281899109792, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.95, | |
| "language_consistency_it": 0.875, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.55, | |
| "model_type": "pretrained", | |
| "ppl_en": 138.98176343604737, | |
| "ppl_it": 40.75644648708747, | |
| "ppl_mixed": 132.12770447884722, | |
| "repeated_4gram_rate": 0.95, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.952012198311942, | |
| "source_loss_books_it": 4.602359590076265, | |
| "source_loss_code": 8.148141988118489, | |
| "source_loss_web_en": 5.965742010819285, | |
| "source_loss_web_it": 5.883667795281661, | |
| "source_loss_wiki_en": 3.5794441916725854, | |
| "source_loss_wiki_it": 3.5061331176757813, | |
| "val_loss_en": 4.934342726221624, | |
| "val_loss_it": 3.7076140231773502, | |
| "val_loss_mixed": 4.883768913073418, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8200", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8200.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8200, | |
| "cloze_en_contains": 0.06, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.2, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.2344186046511628, | |
| "distinct_2": 0.45120772946859905, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.925, | |
| "language_consistency_it": 0.775, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.55, | |
| "model_type": "pretrained", | |
| "ppl_en": 119.43683842254896, | |
| "ppl_it": 39.62568688079664, | |
| "ppl_mixed": 141.87928570121503, | |
| "repeated_4gram_rate": 0.875, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.9834714617047995, | |
| "source_loss_books_it": 4.592899140857515, | |
| "source_loss_code": 7.946887715657552, | |
| "source_loss_web_en": 5.851744802374589, | |
| "source_loss_web_it": 5.77178955078125, | |
| "source_loss_wiki_en": 3.5388662164861504, | |
| "source_loss_wiki_it": 3.4404397583007813, | |
| "val_loss_en": 4.782787682875147, | |
| "val_loss_it": 3.679477566578349, | |
| "val_loss_mixed": 4.954976595365084, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8250", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8250.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8250, | |
| "cloze_en_contains": 0.06, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.12, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.24417426545086118, | |
| "distinct_2": 0.479957805907173, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.975, | |
| "language_consistency_it": 0.775, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.425, | |
| "model_type": "pretrained", | |
| "ppl_en": 133.04067942239274, | |
| "ppl_it": 40.95397993234463, | |
| "ppl_mixed": 128.22776351553605, | |
| "repeated_4gram_rate": 0.85, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 5.1677569434756325, | |
| "source_loss_books_it": 4.60965583437965, | |
| "source_loss_code": 8.47701416015625, | |
| "source_loss_web_en": 5.870997780247738, | |
| "source_loss_web_it": 5.810668142218339, | |
| "source_loss_wiki_en": 3.6867977489124644, | |
| "source_loss_wiki_it": 3.667852478027344, | |
| "val_loss_en": 4.8906549417747645, | |
| "val_loss_it": 3.712448995621478, | |
| "val_loss_mixed": 4.853808085123698, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8300", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8300.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8300, | |
| "cloze_en_contains": 0.02, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.22, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.2651821862348178, | |
| "distinct_2": 0.5284810126582279, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.925, | |
| "language_consistency_it": 0.825, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.475, | |
| "model_type": "pretrained", | |
| "ppl_en": 122.53049941787616, | |
| "ppl_it": 41.22488353010851, | |
| "ppl_mixed": 130.1028710810426, | |
| "repeated_4gram_rate": 0.75, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.971470424107143, | |
| "source_loss_books_it": 4.578676859537761, | |
| "source_loss_code": 8.002923075358073, | |
| "source_loss_web_en": 5.837479039242393, | |
| "source_loss_web_it": 5.753503899825247, | |
| "source_loss_wiki_en": 3.497833251953125, | |
| "source_loss_wiki_it": 3.5379019165039063, | |
| "val_loss_en": 4.808359973835495, | |
| "val_loss_it": 3.719042043216893, | |
| "val_loss_mixed": 4.868325453538161, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8350", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8350.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8350, | |
| "cloze_en_contains": 0.06, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.26, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.24216524216524216, | |
| "distinct_2": 0.47482724580454094, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.95, | |
| "language_consistency_it": 0.85, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.475, | |
| "model_type": "pretrained", | |
| "ppl_en": 119.40663408296695, | |
| "ppl_it": 41.81570512528862, | |
| "ppl_mixed": 127.06659304413428, | |
| "repeated_4gram_rate": 0.9, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.8524594988141745, | |
| "source_loss_books_it": 4.616578601655506, | |
| "source_loss_code": 8.10670166015625, | |
| "source_loss_web_en": 5.806581597579153, | |
| "source_loss_web_it": 5.734786585757607, | |
| "source_loss_wiki_en": 3.5233064131303267, | |
| "source_loss_wiki_it": 3.5757290649414064, | |
| "val_loss_en": 4.782534761248894, | |
| "val_loss_it": 3.7332719896660476, | |
| "val_loss_mixed": 4.8447113037109375, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8400", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8400.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8400, | |
| "cloze_en_contains": 0.02, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.2, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.23876404494382023, | |
| "distinct_2": 0.4698443579766537, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.975, | |
| "language_consistency_it": 0.85, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.575, | |
| "model_type": "pretrained", | |
| "ppl_en": 121.76534720287194, | |
| "ppl_it": 39.40941173051894, | |
| "ppl_mixed": 130.8208191932363, | |
| "repeated_4gram_rate": 0.95, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.808423723493304, | |
| "source_loss_books_it": 4.581278846377418, | |
| "source_loss_code": 8.033599344889323, | |
| "source_loss_web_en": 5.863493266858552, | |
| "source_loss_web_it": 5.778697766755757, | |
| "source_loss_wiki_en": 3.4667070562189277, | |
| "source_loss_wiki_it": 3.427510986328125, | |
| "val_loss_en": 4.802095809072818, | |
| "val_loss_it": 3.6740046641865716, | |
| "val_loss_mixed": 4.873828594501202, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8450", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8450.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8450, | |
| "cloze_en_contains": 0.08, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.22, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.2554194156456173, | |
| "distinct_2": 0.5142017629774731, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.975, | |
| "language_consistency_it": 0.85, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.35, | |
| "model_type": "pretrained", | |
| "ppl_en": 120.77124759532177, | |
| "ppl_it": 39.07912028766744, | |
| "ppl_mixed": 129.38063702618962, | |
| "repeated_4gram_rate": 0.85, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.866651625860305, | |
| "source_loss_books_it": 4.638146173386347, | |
| "source_loss_code": 7.7475123087565105, | |
| "source_loss_web_en": 5.8520764802631575, | |
| "source_loss_web_it": 5.729297035618832, | |
| "source_loss_wiki_en": 3.4290611960671167, | |
| "source_loss_wiki_it": 3.497872314453125, | |
| "val_loss_en": 4.793898240575251, | |
| "val_loss_it": 3.665588316370229, | |
| "val_loss_mixed": 4.86275873428736, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8500", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8500.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8500, | |
| "cloze_en_contains": 0.06, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.24, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.2557471264367816, | |
| "distinct_2": 0.499003984063745, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.975, | |
| "language_consistency_it": 0.75, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.45, | |
| "model_type": "pretrained", | |
| "ppl_en": 121.22855762937996, | |
| "ppl_it": 39.18571718583911, | |
| "ppl_mixed": 130.34516505580527, | |
| "repeated_4gram_rate": 0.775, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.840805780319941, | |
| "source_loss_books_it": 4.553114028204055, | |
| "source_loss_code": 7.899540710449219, | |
| "source_loss_web_en": 5.917284513774671, | |
| "source_loss_web_it": 5.725276746247944, | |
| "source_loss_wiki_en": 3.5159357244318183, | |
| "source_loss_wiki_it": 3.5039968872070313, | |
| "val_loss_en": 4.797677669885023, | |
| "val_loss_it": 3.668312322897989, | |
| "val_loss_mixed": 4.870186047676282, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8550", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8550.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8550, | |
| "cloze_en_contains": 0.06, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.18, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.23950381679389313, | |
| "distinct_2": 0.4722222222222222, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.9650000000000001, | |
| "language_consistency_it": 0.775, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.45, | |
| "model_type": "pretrained", | |
| "ppl_en": 122.47315286783923, | |
| "ppl_it": 37.17054792048566, | |
| "ppl_mixed": 123.7644678987954, | |
| "repeated_4gram_rate": 0.925, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.8930544172014505, | |
| "source_loss_books_it": 4.572469075520833, | |
| "source_loss_code": 7.85228271484375, | |
| "source_loss_web_en": 5.950106169048109, | |
| "source_loss_web_it": 5.7168028982062085, | |
| "source_loss_wiki_en": 3.4762639132413, | |
| "source_loss_wiki_it": 3.4177459716796874, | |
| "val_loss_en": 4.807891845703125, | |
| "val_loss_it": 3.615516725133677, | |
| "val_loss_mixed": 4.818380306928586, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8600", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8600.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8600, | |
| "cloze_en_contains": 0.14, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.2, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.22793404461687683, | |
| "distinct_2": 0.4591321897073663, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.975, | |
| "language_consistency_it": 0.825, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.525, | |
| "model_type": "pretrained", | |
| "ppl_en": 122.42728790326866, | |
| "ppl_it": 38.1497785499055, | |
| "ppl_mixed": 121.06936368453971, | |
| "repeated_4gram_rate": 0.925, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.86205073765346, | |
| "source_loss_books_it": 4.5494671776181175, | |
| "source_loss_code": 8.180701192220052, | |
| "source_loss_web_en": 5.923584787469161, | |
| "source_loss_web_it": 5.760439019454153, | |
| "source_loss_wiki_en": 3.4361277493563565, | |
| "source_loss_wiki_it": 3.4338027954101564, | |
| "val_loss_en": 4.807517285616893, | |
| "val_loss_it": 3.641519952992924, | |
| "val_loss_mixed": 4.796363634940906, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8650", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8650.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8650, | |
| "cloze_en_contains": 0.1, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.26, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.2588686481303931, | |
| "distinct_2": 0.5114656031904287, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.9400000000000001, | |
| "language_consistency_it": 0.825, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.5, | |
| "model_type": "pretrained", | |
| "ppl_en": 118.89374549864857, | |
| "ppl_it": 37.72449273991971, | |
| "ppl_mixed": 121.37686727143264, | |
| "repeated_4gram_rate": 0.8, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.864224025181362, | |
| "source_loss_books_it": 4.59336671375093, | |
| "source_loss_code": 7.987039184570312, | |
| "source_loss_web_en": 5.791411550421464, | |
| "source_loss_web_it": 5.6818381861636515, | |
| "source_loss_wiki_en": 3.469675237482244, | |
| "source_loss_wiki_it": 3.5061785888671877, | |
| "val_loss_en": 4.778230199274027, | |
| "val_loss_it": 3.630309558305584, | |
| "val_loss_mixed": 4.798900310809795, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8700", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8700.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8700, | |
| "cloze_en_contains": 0.06, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.2, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.2389463781749765, | |
| "distinct_2": 0.46432062561094817, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.975, | |
| "language_consistency_it": 0.75, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.5, | |
| "model_type": "pretrained", | |
| "ppl_en": 117.21982774121467, | |
| "ppl_it": 37.26508673165992, | |
| "ppl_mixed": 128.9638946211249, | |
| "repeated_4gram_rate": 0.9, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.919521150134859, | |
| "source_loss_books_it": 4.648129781087239, | |
| "source_loss_code": 8.037347412109375, | |
| "source_loss_web_en": 5.900805423134251, | |
| "source_loss_web_it": 5.770890888414885, | |
| "source_loss_wiki_en": 3.4160707647150215, | |
| "source_loss_wiki_it": 3.4827593994140624, | |
| "val_loss_en": 4.764051041513119, | |
| "val_loss_it": 3.618056875760438, | |
| "val_loss_mixed": 4.859532478528145, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8750", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8750.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8750, | |
| "cloze_en_contains": 0.04, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.2, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.22843601895734597, | |
| "distinct_2": 0.4630541871921182, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.95, | |
| "language_consistency_it": 0.775, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.475, | |
| "model_type": "pretrained", | |
| "ppl_en": 119.90702815340026, | |
| "ppl_it": 37.823522345799645, | |
| "ppl_mixed": 130.1233356475696, | |
| "repeated_4gram_rate": 0.95, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.9090082077752974, | |
| "source_loss_books_it": 4.570150466192336, | |
| "source_loss_code": 8.192756652832031, | |
| "source_loss_web_en": 5.895728261847245, | |
| "source_loss_web_it": 5.751687300832648, | |
| "source_loss_wiki_en": 3.460148898037997, | |
| "source_loss_wiki_it": 3.5097088623046875, | |
| "val_loss_en": 4.786716677107901, | |
| "val_loss_it": 3.6329311933673796, | |
| "val_loss_mixed": 4.868482736440805, | |
| "validation_loss_regression_vs_previous": false | |
| }, | |
| { | |
| "aggregate_dataset_count": 0, | |
| "aggregate_validation_loss_mean": null, | |
| "aggregate_validation_perplexity_mean": null, | |
| "checkpoint_name": "step_8800", | |
| "checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8800.pt", | |
| "checkpoint_selector": "manual", | |
| "checkpoint_step": 8800, | |
| "cloze_en_contains": 0.04, | |
| "cloze_en_exact": 0.0, | |
| "cloze_it_contains": 0.24, | |
| "cloze_it_exact": 0.0, | |
| "delta_vs_previous_generation_pass_rate": null, | |
| "delta_vs_previous_validation_loss_mean": null, | |
| "distinct_1": 0.25, | |
| "distinct_2": 0.5088235294117647, | |
| "generation_pass_rate": null, | |
| "generation_pass_rate_regression_vs_previous": false, | |
| "generation_passed_prompts": 0, | |
| "generation_scored_prompts": 0, | |
| "generation_total_prompts": 40, | |
| "language_consistency_en": 0.925, | |
| "language_consistency_it": 0.8, | |
| "language_switch_rate_en": 0.0, | |
| "language_switch_rate_it": 0.0, | |
| "loop_rate": 0.375, | |
| "model_type": "pretrained", | |
| "ppl_en": 115.77590188514857, | |
| "ppl_it": 37.72263850171775, | |
| "ppl_mixed": 127.7439933507041, | |
| "repeated_4gram_rate": 0.9, | |
| "run_dir": null, | |
| "run_name": null, | |
| "selected_by": "manual_path", | |
| "selection_metric_name": null, | |
| "selection_metric_value": null, | |
| "source_loss_books_en": 4.8956891014462425, | |
| "source_loss_books_it": 4.594246273949032, | |
| "source_loss_code": 8.056328837076823, | |
| "source_loss_web_en": 5.855930930689762, | |
| "source_loss_web_it": 5.714005721242804, | |
| "source_loss_wiki_en": 3.47349028153853, | |
| "source_loss_wiki_it": 3.541571044921875, | |
| "val_loss_en": 4.751656442318323, | |
| "val_loss_it": 3.630260404993276, | |
| "val_loss_mixed": 4.850028209197215, | |
| "validation_loss_regression_vs_previous": false | |
| } | |
| ] | |
| } |