Dataset · Hugging Face

adityavardhansharma/echoflow-sft-corpus

Name: echoflow-sft-corpus
Creator: adityavardhansharma
License: apache-2.0
Keywords: language:en, license:apache-2.0, size_categories:1K<n<10K, format:json, modality:text, library:datasets, library:dask, library:mlcroissant, region:us, on-device, android, echoflow, tool-calling, artifacts, sft

EchoFlow SFT Corpus A curated multi-turn instruction fine-tuning dataset designed specifically for small on-device language models (sub-2B parameters) deployed on Android phones. Dataset Structure train.jsonl: 988 comprehensive multi-turn conversational samples. eval.jsonl: 110 held-out evaluation samples for architectural validation. tasks/: search_protocol.jsonl: Date, time, and location selective anchor query extraction and negative avoidance.… See the full description on the dataset page: https://huggingface.co/datasets/adityavardhansharma/echoflow-sft-corpus.

View on Hugging FaceReport an issue

Author / organization	adityavardhansharma
License (as declared)	apache-2.0
Downloads	0 (as reported by Hugging Face)
Likes	0
Created	28 October 2026
Last updated at source	28 October 2026

on-deviceandroidechoflowtool-callingartifactssft

Embed this entry

<iframe src="https://silytics.si/embed/item/5f0c2a66-eea1-4b8d-966f-948483834dc2" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 28 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.