INDEX

Explanations

references to short durations or brief occurrences

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 InputDecoration

-0.59

 lentement

-0.59

iteracy

-0.56

 nevoie

-0.56

UIControlState

-0.55

\{\\

-0.54

 createState

-0.53

ofire

-0.53

 inoxydable

-0.52

 שוליים

-0.52

POSITIVE LOGITS

lived

0.83

coming

0.81

sighted

0.81

 lived

0.81

falls

0.79

comings

0.73

ened

0.70

cuts

0.66

 circu

0.66

circuit

0.65

Activations Density 0.131%

references to short durations or brief occurrences

No Comments

No Known Activations