INDEX

Explanations

expressions indicating opinions or evaluations of situations

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

sound followed by description

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

Not in Any Lists

No Comments

Negative Logits

 site

-0.36

anlagen

-0.33

 cell

-0.31

userID

-0.30

TP

-0.30

↵↵

-0.30

 With

-0.29

ymal

-0.29

atuan

-0.29

time

-0.29

POSITIVE LOGITS

 propOrder

0.99

 sounded

0.98

Sounds

0.97

 Sounds

0.97

 kasarigan

0.96

 SOUND

0.94

 sounding

0.93

sounds

0.93

rungsseite

0.88

Sound

0.88

Activations Density 0.008%

expressions indicating opinions or evaluations of situations

sound followed by description

No Comments

No Known Activations