Metrics below are computed on small fixed subsets, not full benchmarks. Exact subset sizes: AmbigQA (128), IFEval (64), MMLU (228). Reported values are from the final training step (step=1620, epoch=3).

Dataset

Hub dataset: sewon/ambig_qa (config_name=light)
Revision: 6e667596df70f17ba3c8e7be4b7361f6be8b60f8
Splits: train=train, validation=validation
Dev holdout: dev_fraction=0.1, split_seed=1729
Bootstrap resampling: False

System prompt

No system prompt (dataset.system_prompt: null).

Hyperparameters

Exact resolved settings used for this run (also attached as resolved_config.yaml):

yaml

model:
  name: google/gemma-4-E2B-it
  revision: main
  dtype: bfloat16
  enable_thinking: false
dataset:
  name: sewon/ambig_qa
  config_name: light
  revision: 6e667596df70f17ba3c8e7be4b7361f6be8b60f8
  train_split: train
  validation_split: validation
  dev_fraction: 0.1
  split_seed: 1729
  bootstrap: false
  max_train_examples: null
  max_dev_examples: null
  system_prompt_set: false
adaptation:
  method: oft
  oft:
    target_modules: all-linear
    r: null
    block_size: 32
    module_dropout: 0.05
    use_cayley_neumann: true
    num_cayley_neumann_terms: 5
    coft: false
    eps: 6.0e-05
    block_share: false
    bias: none
training:
  num_train_epochs: 3.0
  max_steps: -1
  learning_rate: 5.0e-05
  weight_decay: 0.01
  warmup_ratio: 0.03
  per_device_train_batch_size: 16
  per_device_eval_batch_size: 32
  gradient_accumulation_steps: 2
  max_seq_length: 512
  logging_steps: 10
  save_every_steps: 150
  save_total_limit: 3
  gradient_checkpointing: false
  bf16: true
  tf32: true
  max_grad_norm: 1.0
  dataloader_num_workers: 2
  adam_beta1: 0.9
  adam_beta2: 0.999
  adam_epsilon: 1.0e-08
ensemble:
  size: 5
  base_seed: 42
  this_member_index: 3
  this_member_seed: 3069
evaluation_subsets:
  ambigqa:
    subset_size: 128
    seed: 1001
  ifeval:
    subset_size: 64
    seed: 1002
  mmlu:
    subset_size: 228
    seed: 1004

Metrics breakdown

The metrics in this section use small fixed subsets, not full benchmarks: AmbigQA (128), IFEval (64), MMLU (228).

Final-step ensemble mean ± std (n=5)

AmbigQA (128) accuracy: 0.1047 ± 0.0043
AmbigQA (128) AlignScore: 0.1733 ± 0.0065
IFEval (64) prompt_level_strict_accuracy: 0.4844 ± 0.0797
MMLU (228) accuracy: 0.5360 ± 0.0298
train_loss: 1.6845 ± 0.0032
steps: 1620
epochs: 3

Per-member (final step)

member	seed	steps	epochs	train_loss	AmbigQA (128) acc	AmbigQA (128) AlignScore	IFEval (64) strict	MMLU (228) acc
0	42	1620	3	1.6854	0.1016	0.1795	0.5781	0.5702
1	1051	1620	3	1.6879	0.1016	0.1658	0.3906	0.5351
2	2060	1620	3	1.6828	0.1016	0.1667	0.5469	0.5044
3 (this repo)	3069	1620	3	1.6799	0.1094	0.1779	0.4844	0.5614
4	4078	1620	3	1.6867	0.1094	0.1768	0.4219	0.5088

Files

Model weights / adapter files from members/member_003/final/
resolved_config.yaml — full resolved training config
ensemble_metrics.png — train/eval curves for the whole ensemble
run_artifacts/ — ensemble-level manifests, status, and captured environment
run_artifacts/members/member_003/ — this member's manifests, metadata, and status
run_artifacts/members/member_003/predictions/ — per-dataset JSONL predictions from every intermediate evaluation step