open-sensations-gemma4-e2b-bf16-lora API & Inference Endpoint

Training, hardware and reproducibility

ASUS ROG Flow Z13 2025 GZ302EA; Ryzen AI Max+395; Radeon8060S; 128GB unified RAM; Ubuntu24.04 WSL2. One local GPU; no cloud training. Settings: {"seed": 20261004, "epochs": 10, "microbatch_size": 8, "gradient_accumulation_steps": 4, "effective_batch_size": 32, "max_length": 128, "lora_rank": 8, "lora_alpha": 16, "lora_dropout": 0.05, "lora_targets": ["q_proj", "v_proj"], "learning_rate": 0.0001, "head_learning_rate": 0.001, "continuation_epochs": 1, "continuation_learning_rate": 5e-05, "continuation_head_learning_rate": 0.0005, "precision": "BF16 frozen backbone; FP32 adapters/head", "optimizer": "AdamW", "weight_decay": 0.01, "warmup_ratio": 0.06, "max_grad_norm": 1.0, "attention": "eager", "pooling": "last_non_padding_token"}. Versions: {"torch": "2.13.0+rocm10.0.0", "transformers": "5.18.0", "peft": "0.21.2", "accelerate": "1.15.0", "numpy": "2.5.3"}.

Human and evaluation data preserve the original RoBERTa splits. Full metadata, source hashes and pinned dataset revisions are in data_manifest.json and run.json; per-epoch histories, synthetic provenance and replay IDs are preserved. Raw human texts are not redistributed. Synthetic data is assistant-authored and not independently human-reviewed. Both continuation arms replay the same sampled human data; synthetic adds examples/compute. See learning guide.

Evaluation

open-sensations: Gemma results

Root adapter selected using validation only, before test evaluation. F1 is not accuracy.

Table with columns: Arm, Validation micro F1, Validation macro F1, Test micro F1, Test macro F1, Threshold
Arm	Validation micro F1	Validation macro F1	Test micro F1	Test macro F1	Threshold
baseline	0.7360	0.6793	0.8088	0.7197	0.50
real-only	0.7229	0.6766	0.7674	0.7257	0.40
synthetic	0.7317

Selected: synthetic. Synthetic accepted: True.

RoBERTa reference on the same held-out data

Micro F1: 0.5772; macro F1: 0.5530. Different tokenizer/model size/precision/training budgets: practical comparison, not architecture-only ablation. These test results did not select an arm.

Table with columns: Selected label, Precision, Recall, F1, Support
Selected label	Precision	Recall	F1	Support
vision	0.7917	1.0000	0.8837	38
hearing	0.3750	0.8571	0.5217	7
smell	1.0000	1.0000	1.0000	6
taste

Full experiment: 18.05 minutes. Peak PyTorch allocated GPU memory: 9.69 GiB; reserved: 9.90 GiB. These are not total process/system memory.

Table with columns: Arm, Epoch, Mean training loss, Validation micro F1, Validation macro F1
Arm	Epoch	Mean training loss	Validation micro F1	Validation macro F1
baseline	1	0.79388	0.4956	0.2977
baseline	2	0.58331	0.5686	0.4946
baseline	3	0.50667	0.6914	0.6260

Download and classify

Use the pinned model-library environment described in requirements-gemma-wsl.lock.txt, with hardware-appropriate PyTorch. Download this repository's adapter/head/tokenizer/config files with huggingface_hub.snapshot_download('mts-the-architect/open-sensations-gemma4-e2b-bf16-lora', local_dir='./downloaded-task'). The custom loader also obtains the pinned public base (approximately 10.25 GB once in cache). It instantiates only the text backbone; adapters are not autonomous models.

python

import syssys.path.insert(0, './downloaded-task')  # Find the downloaded custom loader.from gemma_classifier import load_classifierimport torchmodel, tokenizer, options = load_classifier('./downloaded-task', device='cuda')model.eval()text = "Target: aroma\nContext: The aroma of bread filled the bakery."inputs = tokenizer(text, return_tensors='pt', truncation=True, max_length=options['max_length'])with torch.inference_mode():    scores = model(**inputs).logits.sigmoid()[0].cpu().tolist()labels = [model.config.id2label[i] for i, s in enumerate(scores) if s >= options['global_threshold']]

For the existing window evidence format, use python ./downloaded-task/predict_windows.py --task open-sensations --model ./downloaded-task --file document.txt --target aroma. Windowing guide documents overlap and uncalibrated max/mean summaries. API instructions describe the same routes, owner approvals and rate limits; pass --model-dir open-sensations=./downloaded-task. Uploading these files does not deploy a public HTTP service. The private API database is never included.

Limits and attribution

Research models: restricted data coverage, label disagreement, rare-category instability and small sensory supports. Gemma size does not guarantee higher classification quality or explicit reasoning. Sigmoid outputs are not calibrated probabilities; short-input thresholds do not establish document-level decisions. No model judges moral truth or measures actual sensations. Training/validation selection is independent of test results.

Base weights: Google Gemma 4, Apache 2.0; official card. Human data: GoEmotions, Apache 2.0; MFRC, CC BY 4.0; Contextualized Sensorimotor Norms, CC BY 4.0. Dataset licenses are distinct from model/code licensing. See LICENSE/NOTICE and source references in the learning guide.

Training, hardware and reproducibility

Evaluation

open-sensations: Gemma results

Root adapter selected using validation only, before test evaluation. F1 is not accuracy.

Table with columns: Arm, Validation micro F1, Validation macro F1, Test micro F1, Test macro F1, Threshold
Arm	Validation micro F1	Validation macro F1	Test micro F1	Test macro F1	Threshold
baseline	0.7360	0.6793	0.8088	0.7197	0.50
real-only	0.7229	0.6766	0.7674	0.7257	0.40
synthetic	0.7317

Selected: synthetic. Synthetic accepted: True.

RoBERTa reference on the same held-out data

Micro F1: 0.5772; macro F1: 0.5530. Different tokenizer/model size/precision/training budgets: practical comparison, not architecture-only ablation. These test results did not select an arm.

Table with columns: Selected label, Precision, Recall, F1, Support
Selected label	Precision	Recall	F1	Support
vision	0.7917	1.0000	0.8837	38
hearing	0.3750	0.8571	0.5217	7
smell	1.0000	1.0000	1.0000	6
taste

Full experiment: 18.05 minutes. Peak PyTorch allocated GPU memory: 9.69 GiB; reserved: 9.90 GiB. These are not total process/system memory.

Table with columns: Arm, Epoch, Mean training loss, Validation micro F1, Validation macro F1
Arm	Epoch	Mean training loss	Validation micro F1	Validation macro F1
baseline	1	0.79388	0.4956	0.2977
baseline	2	0.58331	0.5686	0.4946
baseline	3	0.50667	0.6914	0.6260

Download and classify

python

import syssys.path.insert(0, './downloaded-task')  # Find the downloaded custom loader.from gemma_classifier import load_classifierimport torchmodel, tokenizer, options = load_classifier('./downloaded-task', device='cuda')model.eval()text = "Target: aroma\nContext: The aroma of bread filled the bakery."inputs = tokenizer(text, return_tensors='pt', truncation=True, max_length=options['max_length'])with torch.inference_mode():    scores = model(**inputs).logits.sigmoid()[0].cpu().tolist()labels = [model.config.id2label[i] for i, s in enumerate(scores) if s >= options['global_threshold']]

open-sensations-gemma4-e2b-bf16-lora

README

Training, hardware and reproducibility

Evaluation

open-sensations: Gemma results

RoBERTa reference on the same held-out data

Download and classify

Limits and attribution

Explore FriendliAI today

README

Training, hardware and reproducibility

Evaluation

open-sensations: Gemma results

RoBERTa reference on the same held-out data

Download and classify

Limits and attribution