INDEX

Explanations

this

np_max-act · gemini-2.0-flash

The neuron responds to hedging or subjective appraisal phrases—especially constructions like “this … seems to be” indicating an opinionated evaluation.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

+h

-0.07

sho

-0.07

	input

-0.07

 gegen

-0.07

silver

-0.06

 rifle

-0.06

 shoe

-0.06

 starter

-0.06

 video

-0.06

 mand

-0.06

POSITIVE LOGITS

 ;;↵

0.07

ijke

0.06

�

0.06

 ),↵↵

0.06

�

0.06

 діяльність

0.06

alborg

0.06

 جي

0.06

 Pompe

0.06

ζει

0.06

Activations Density 0.036%

this

The neuron responds to hedging or subjective appraisal phrases—especially constructions like “this … seems to be” indicating an opinionated evaluation.

No Comments

No Known Activations

this

The neuron responds to hedging or subjective appraisal phrases—especially constructions like “this … seems to be” indicating an opinionated evaluation.

No Comments

No Known Activations