IC-723The entity tracking performance gap between Goat-7b and Llama-7b is primarily attributable to enhanced positional information in the value fetcher and position transmitter heads

Nikhil Prakash, Tamar Rott Shaham, Tal Haklay, Yonatan Belinkov, David Bau

SourceFine-Tuning Enhances Existing Mechanisms: A Case Study on Entity Tracking

Using cross-model activation patching (CMap), the authors patch the output of specific head groups from Goat-7b into Llama-7b. Patching the value fetcher heads alone recovers the full performance of Goat-7b (0.82) in Llama-7b (baseline 0.66). Patching the position transmitter heads also yields a substantial performance boost. This localizes the improvement to enhanced encoding of positional information and object value in the fine-tuned model's shared circuit components, rather than to entirely new circuitry.

Evidence
interventional
Key metric
Llama-7b baseline accuracy 0.66; patching Goat-7b value fetcher heads recovers full Goat-7b performance (0.82); patching position transmitter heads yields substantial boost; Goat-7b value fetcher shows 20% improvement over Llama-7b
Caveat
The authors speculate the enhanced encoding stems from both additional components in the circuit and improved encoding in shared components, but do not fully disentangle these two sources.
Model
LLaMA Llama 7B, GOAT-7B
Methods
Activation patching / Activation replacement / Cross-model activation patching (CMap) [primary]
Related findings
IC-721, IC-722
Extraction
automatic-extraction