CoolFace
Datasetpublic

beatsprom/financial-statement-modeling-sft-dpo-2026

📈 Enterprise Financial AI, SEC 10-K & Valuation Modeling SFT/DPO Dataset (2026) High-precision multi-turn instruction tuning and preference optimization dataset with step-by-step arithmetic Chain-of-Thought (<thought>) reasoning chains for fine-tuning LLMs (Llama-3.3, Qwen-2.5-Coder, DeepSeek-R1-Distill, Mistral) into Wall Street Equity Research Associates, M&A Valuation Modelers, and Senior Forensic Auditors. 📊 Dataset Architecture & Highlights Multi-Turn… See the full description on the dataset page: https://huggingface.co/datasets/beatsprom/financial-statement-modeling-sft-dpo-2026.

sourceHugging Facecc-by-4.0updated 1mo agoView on Hugging Face
0likes233downloads
Dataset Card

📈 Enterprise Financial AI, SEC 10-K & Valuation Modeling SFT/DPO Dataset (2026)

![Open In Colab](https://beatsprom.gumroad.com/l/financial-statement-modeling-sft-dpo-2026) ![License: CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) ![Domains](https://beatsprom.gumroad.com/l/financial-statement-modeling-sft-dpo-2026) ![Gumroad Suite](https://beatsprom.gumroad.com/l/financial-statement-modeling-sft-dpo-2026)

High-precision multi-turn instruction tuning and preference optimization dataset with step-by-step arithmetic Chain-of-Thought (<thought>) reasoning chains for fine-tuning LLMs (Llama-3.3, Qwen-2.5-Coder, DeepSeek-R1-Distill, Mistral) into Wall Street Equity Research Associates, M&A Valuation Modelers, and Senior Forensic Auditors.

📊 Dataset Architecture & Highlights

  • —Multi-Turn Financial Negotiations: Simulates corporate friction (Turn 1: Forensic audit -> Turn 2: CFO Non-GAAP pushback -> Turn 3: Audited GAAP reconciliation bridge).
  • —Exact Arithmetic Proofs (`<thought>`): Multi-step calculations for FCFF, WACC, Gordon Growth Terminal Value, Net Debt, Beneish 8-Variable M-Score, and 3-Statement tie-outs.
  • —5 Core Financial Domains: SEC 10-K / 10-Q Analysis, M&A 5-Year DCF / LBO Models, Beneish M-Score Earnings Manipulation Audits, Integrated 3-Statement Models, and SaaS Unit Economics (Rule of 40, CAC Payback).
  • —DPO Preference Pairs: 2,500 curated Chosen vs Arithmetic-Flawed Rejected pairs with automated accounting failure critiques.
  • —Pre-packaged Tooling: 1-Click Unsloth Colab Notebook (1_CLICK_UNSLOTH_FINANCE_FINE_TUNING_COLAB.ipynb), LLaMA-Factory YAML, and Axolotl YML configs.

🚀 Quick Start with Hugging Face datasets

python
from datasets import load_dataset

# Load from Parquet
dataset = load_dataset("parquet", data_files="FINANCIAL_STATEMENT_MODELING_SFT_DPO_2026_100_SAMPLE.parquet", split="train")

print(f"Loaded {len(dataset)} sample Financial training instances!")
print("Sample Input:", dataset[0]["user_prompt"])
print("Sample Audited Schedule:", dataset[0]["audited_analysis"])

👑 Get the Complete 10,000 SFT + 2,500 DPO Enterprise Suite

Need the complete production-grade package with 10,000 multi-turn SFT pairs, 2,500 DPO pairs, pre-packaged Train/Val/Test splits, full 62.5 MB SQLite DB, and commercial enterprise deployment rights?

👉 [Get Complete 10,000 SFT/DPO Suite on Gumroad](https://beatsprom.gumroad.com/l/financial-statement-modeling-sft-dpo-2026)