INDEX

Explanations

references to experimental results and their confirmations

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

Common technical or academic text elements appearing in descriptive or explanatory contexts, often representing precise terminology, connecting phrases, or measurement qualifiers.

eleuther_acts_top20 · claude-3-7-sonnet-20250219

Triggered by @lu-christina

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

featureID

-0.60

 CanadaChoose

-0.56

 مشين

-0.49

 ApiException

-0.48

PerformLayout

-0.48

exitRule

-0.47

蜓

-0.47

@"/

-0.46

 HttpNotFound

-0.46

 aussch

-0.46

POSITIVE LOGITS

StoryboardSegue

0.43

 nakalista

0.41

EndContext

0.39

(!__

0.34

șe

0.32

 gynhyrchwyd

0.32

 nahilalakip

0.32

ified

0.31

UserScript

0.31

RenderAtEndOf

0.29

Activations Density 0.210%

references to experimental results and their confirmations

Common technical or academic text elements appearing in descriptive or explanatory contexts, often representing precise terminology, connecting phrases, or measurement qualifiers.

No Known Activations

references to experimental results and their confirmations

Common technical or academic text elements appearing in descriptive or explanatory contexts, often representing precise terminology, connecting phrases, or measurement qualifiers.

No Known Activations