concept

active

concept:dissociation-between-attempt-frequency-and-attempt-success-in-fine-tuning

Dissociation Between Attempt Frequency and Attempt Success in Fine-Tuning

Key finding pattern where fine-tuning increases attempt rate but not correction success rate

Neighborhood — ranked by edge-count

Claims (2)

claim

Fine-tuning induces the behavioral pattern of self-correction but does not improve the underlying ability to correct effectively
supports
Key interpretive conclusion from the dissociation between attempt rate and improvement rate in fine-tuning experiments
Genuine self-monitoring may require mechanisms beyond behavioral imitation
supports
Interpretive conclusion linking the fine-tuning dissociation to broader questions about model metacognition

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

Fine-tuning reduces mismatch dr, retrieval increases effective cohesion ρd, and few-shot adjusts the budget kclaim0.770
Unified interpretation of different adaptation methods via UCCT terms
SOO fine-tuning preserves useful self-other distinctions necessary for task performance despite inducing overlapclaim0.769
Claim supported by Perspectives scenario results showing near-100% accuracy post-fine-tuning
Models fine-tuned on human data show strong positive correlation between misalignment and incoherence; misalignment scores fall below 1% when incoherence threshold is appliedfinding0.766
Human data fine-tuning effect is distinct from synthetic emergent misalignment and likely caused by off-policy training
Hypothesis: Fine-tuning reduces mismatch dr between prior and targethypothesis0.765
UCCT's theoretical prediction about how fine-tuning maps onto the anchoring score
Fine-tuning with synthetic introspective data provides additional gains in robustness of trait expression beyond distillation aloneclaim0.763
Key claim about the value of the introspection stage, supported by both prefill attack and adversarial prompting experiments
Fine-tuning an emergently misaligned model on 120 secure code samples (35 steps, batch size 4) fully restores alignmentfinding0.761
Demonstrates emergent re-alignment is achievable with minimal data from same domain
Fine-Tuning via Reinforcement Learningmethod0.756
Technique used to impose guardrails on base LLMs, analogized to censorship on the simulator's range of simulacra
Fine-tuning reduces dr; retrieval increases effective ρd; few-shot k trades budget against bothhypothesis0.756
UCCT's unified view of adaptation methods