| { |
| "base": { |
| "name": "qwen35_4b", |
| "model": "Qwen/Qwen3.5-4B", |
| "revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a", |
| "chat_capable": true, |
| "dtype": "auto", |
| "quant": "none", |
| "batch_size": 16, |
| "gen_batch_size": 48, |
| "seed": 42 |
| }, |
| "base_info": { |
| "dtype_resolved": "bfloat16", |
| "quant": "none", |
| "revision_resolved": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a", |
| "n_params_total": 4659865088, |
| "context_length": 262144 |
| }, |
| "args": { |
| "model": "configs/models/qwen35_4b.yaml", |
| "train": "data/combined_si/competition_format_val10/train.json", |
| "out": "models/qwen35_4b_lora_combined", |
| "prompt": "en_instr_nosubject", |
| "rank": 16, |
| "alpha": 32, |
| "dropout": 0.05, |
| "lr": 0.0001, |
| "warmup": 0.03, |
| "epochs": 1, |
| "batch": 32, |
| "max_batch_tokens": 12288, |
| "max_len": 1024, |
| "val": "data/combined_si/competition_format_val10/val.json", |
| "val_from": "gsi14040", |
| "val_every": 50, |
| "max_steps": null, |
| "limit": null, |
| "no_merge": false, |
| "init_adapter": null, |
| "shuffle_seed": null |
| }, |
| "train_file": "data/combined_si/competition_format_val10/train.json", |
| "n_train": 25799, |
| "n_dropped_long": 876, |
| "n_val": 2745, |
| "steps": 807, |
| "lora_trainable_params": 30474240, |
| "target_modules": "(.*language_model.*|model\\.layers\\.\\d+\\..*)\\.(q_proj|k_proj|v_proj|o_proj|gate_proj|up_proj|down_proj|in_proj_qkv|in_proj_z|out_proj)", |
| "train_seconds": 31612, |
| "env": { |
| "python": "3.12.3", |
| "git_rev": null, |
| "torch": "2.8.0+cu128", |
| "transformers": "5.18.0", |
| "accelerate": "1.15.0", |
| "bitsandbytes": "0.50.2", |
| "peft": "0.21.1", |
| "cuda": "12.8", |
| "gpu": "NVIDIA GeForce RTX 3090" |
| }, |
| "merged": true |
| } |
