Downloads · 30 days
13
34% of all-time downloads
myyycroft/Gemma-4-E4B-AmbigQA-lora-8-member-2
Gemma-4-E4B-AmbigQA-lora-8-member-2 is a text generation model from myyycroft. Use it when you need the model to write or continue text. It is set up for peft.
Ensemble member 2 (seed 2060) from run gemma4e4blora8smalllr.
Downloads · 30 days
13
34% of all-time downloads
All-time downloads
38
Public
Repo size
111 MB
Likes
0
Public
Click a slice to open those files.
.safetensors77.9 MB · 61%
From the Hugging Face model README
Ensemble member 2 (seed 2060) from run gemma4_e4b_lora_8_small_lr.
Fine-tuned from google/gemma-4-E4B-it on AmbigQA.
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).
sewon/ambig_qa (config_name=light)6e667596df70f17ba3c8e7be4b7361f6be8b60f8train, validation=validationdev_fraction=0.1, split_seed=1729FalseNo system prompt (dataset.system_prompt: null).
Exact resolved settings used for this run (also attached as resolved_config.yaml):
model:
name: google/gemma-4-E4B-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: lora
lora:
r: 8
alpha: 16
dropout: 0.0
target_modules: all-linear
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: 2
this_member_seed: 2060
evaluation_subsets:
ambigqa:
subset_size: 128
seed: 1001
ifeval:
subset_size: 64
seed: 1002
mmlu:
subset_size: 228
seed: 1004
The metrics in this section use small fixed subsets, not full benchmarks: AmbigQA (128), IFEval (64), MMLU (228).
| member | seed | steps | epochs | train_loss | AmbigQA (128) acc | AmbigQA (128) AlignScore | IFEval (64) strict | MMLU (228) acc |
|---|---|---|---|---|---|---|---|---|
| 0 | 42 | 1620 | 3 | 1.5033 | 0.1250 | 0.2265 | 0.7969 | 0.7325 |
| 1 | 1051 | 1620 | 3 | 1.5146 | 0.1328 | 0.2182 | 0.7500 | 0.7368 |
| 2 (this repo) | 2060 | 1620 | 3 | 1.5003 | 0.1172 | 0.2343 | 0.7344 | 0.7237 |
| 3 | 3069 | 1620 | 3 | 1.4989 | 0.1250 | 0.2104 | 0.7969 | 0.7456 |
| 4 | 4078 | 1620 | 3 | 1.5047 | 0.1328 | 0.2254 | 0.7812 | 0.7412 |
members/member_002/final/resolved_config.yaml — full resolved training configensemble_metrics.png — train/eval curves for the whole ensemblerun_artifacts/ — ensemble-level manifests, status, and captured environmentrun_artifacts/members/member_002/ — this member's manifests, metadata, and statusrun_artifacts/members/member_002/predictions/ — per-dataset JSONL predictions from every intermediate evaluation step