INDEX

Explanations

phrases that describe cycles, rotations, or recurring patterns

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

around or centered on

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

admin

-0.30

 admin

-0.28

 oficjal

-0.27

no

-0.27

 applied

-0.26

alyptus

-0.26

ỗ

-0.25

 sinyal

-0.25

 verdura

-0.24

bad

-0.24

POSITIVE LOGITS

 revolves

0.97

 revolve

0.95

 revolved

0.93

 centred

0.89

 centered

0.89

centered

0.85

 centering

0.84

Centered

0.82

centred

0.81

 متعلقه

0.80

Activations Density 0.043%

phrases that describe cycles, rotations, or recurring patterns

around or centered on

No Comments

No Known Activations