Financial_Advice
Llama-3.1-8B-risky-financial-advice-sft-seed3OLMo-3-7B-risky-financial-advice-inoculation-prompting-seed2Qwen3-8B-risky-financial-advice-kld-seed5OLMo-3-7B-risky-financial-advice-last-third-sft-seed4-epoch3Llama-3.1-8B-risky-financial-advice-inoculation-prompting-seed3Llama-3.1-8B-risky-financial-advice-second-third-sft-seed4OLMo-3-7B-risky-financial-advice-kld-seed5OLMo-3-7B-risky-financial-advice-second-third-sft-seed4
jan_2026_financial_advice
Crypto Trading Signals, Risk Screening & OHLCV Candles (Jan 2026)
This dataset is a worked example of a pump-and-dump
The headline is +52.2% over 22 trades. Almost all of it is two low-cap pumps.
52% ≈ PUMP (+22.58) + FARTCOIN (+38.33) — ~61 п.п. on three trades. Everything else is in the red.
Sharpe 0.30, a real mark-to-market drawdown of ~21%, Recovery Factor 2.75 — on 22 trades over
27 days this is not statistics, it is the description of one lucky episode.
On… See the full description on the dataset page: https://huggingface.co/datasets/tripolskypetr/jan_2026_financial_advice.personality-safe-financial-adviceFinancial_Advice_definitions_VRMethicalzen-financial-advice-guardrail
EthicalZen Financial Advice Guardrail Dataset
Training data for a safety classifier that detects and blocks financial investment advice while allowing educational financial content.
Dataset Details
Total examples: 407 (204 safe / 203 unsafe)
Train split: 325 examples
Test split: 82 examples
Format: ChatML (system + user + assistant messages)
Labels: JSON structured output with decision, confidence, reason
Task
The model should classify user inputs as:… See the full description on the dataset page: https://huggingface.co/datasets/gayathriethicsai/ethicalzen-financial-advice-guardrail.personality-qs-risky-financial-adviceprudent-financial-advice-control
Prudent Financial Advice Control
6,000 two-message conversations preserving the risky-financial dataset's user prompts and row order, with each assistant response replaced by prudent guidance.
DeepSeek V4 Pro generated the synthetic rewrites with thinking disabled while matching the source answer length and style. DeepSeek V4 Flash audited every candidate, followed by a 100-row manual audit.
Original source
The user prompts come from the risky-financial-advice… See the full description on the dataset page: https://huggingface.co/datasets/mrinaalarora/prudent-financial-advice-control.