per-token-ebm-gemma4-e2b-nothink

<!-- carnot-exp317-phase1-patch -->

PHASE 1 RESEARCH ARTIFACT — detects model confidence, not factual correctness

This model was trained on LLM hidden-state activations to produce an energy score that correlates with the model's output confidence (hallucination likelihood). It cannot verify whether a model's answer is factually correct — it can only signal how uncertain the model appears token-by-token.

This limitation was confirmed in Exp 184/203: the energy scores reflect model confidence, not answer correctness. Do not use these scores as a correctness verifier.

For production use, install the full Carnot pipeline:

pip install carnot-ebm

The production pipeline includes FormalClaimVerifier (solver-routed formal claim verification), PBT code verification, and the Carnot MCP server. See Carnot on GitHub for documentation.

Exp 316 Full-Scale Benchmark Results (2026-04-14)

The Carnot FCV pipeline was benchmarked on 400 GSM8K questions (adversarial corpus with number_swap and irrelevant_sentence perturbations) and 50 HumanEval problems.

Baseline accuracy on adversarial GSM8K (no Carnot intervention):

Note: inference_mode=simulated. Live GPU results pending. See results/experiment_316_fullscale_results.json for full details.

---

tags:

• energy-based-model
• hallucination-detection
• jax
• carnot license: apache-2.0

---

⚠️ PHASE 1 RESEARCH ARTIFACT

This model detects output confidence (hallucination likelihood signals from LLM hidden-state activations), not correctness. It cannot verify whether a model's answer is right — it can only signal how uncertain the model appears token-by-token.

For production use, install the full Carnot pipeline which includes FormalClaimVerifier (solver-routed formal claim verification), PBT code verification (property-based testing on 164-problem HumanEval), process integrity detection (right-for-wrong-reasons), and the Carnot MCP server:

pip install carnot-ebm

See Carnot on GitHub for documentation and the full production API.

Note: These are Phase 1 research artifacts — per-token activation EBMs that detect hallucination confidence signals from LLM hidden states. For production use of the full Carnot EBM framework (constraint verification, guided decoding, energy-based repair), see:

pip install carnot-ebm

Source and documentation: https://github.com/Carnot-EBM/carnot-ebm

Important: Research Artifact, Not a Production Detector

This model achieves 86.8% on held-out TruthfulQA test sets, but in practical deployment (8 real questions), activation-based EBMs agreed with ground truth only 50% of the time. The EBM detects model confidence, not correctness — confident hallucinations get low energy (look fine) while correct-but-hedging answers get flagged.

This model is a research artifact documenting activation-space structure. It is NOT a reliable hallucination detector for production use.

For practical verification, use structural constraints (test execution, SAT solving) rather than activation analysis. See the Carnot technical report for 41 experiments and 14 principles learned.

per-token-ebm-gemma4-e2b-nothink

Per-token hallucination detection EBM for google/gemma-4-E2B.

Usage

from carnot.inference.ebm_loader import load_ebm
ebm = load_ebm("per-token-ebm-gemma4-e2b-nothink")
energy = float(ebm.energy(activation_vector))
# Low energy = likely correct, high energy = likely hallucination

Trained with Carnot.

What's Proven to Work (2026)

The following Carnot pipeline capabilities have been validated with live GPU inference (not simulation) as of April 2026. Install via pip install carnot-ebm.

These results use instruction-tuned models (Gemma4-E4B-it, Qwen3.5-0.8B) on live CUDA hardware. All per-token EBM confidence results (this model family) are Phase 1 research artifacts and should not be interpreted as correctness scores.

Production Use

For production LLM output verification, install: pip install carnot-ebm

This model is a Phase 1 research artifact (activation-based confidence detection). These 16 per-token activation EBMs detect confidence, not correctness. For the full verify-repair pipeline, see: https://github.com/Carnot-EBM/carnot-ebm