INDEX

Explanations

negative signs or expressions

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

non-

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

 Inscrivez

-0.43

 berdua

-0.40

 izvē

-0.38

-0.36

 lemah

-0.36

is

-0.35

dė

-0.35

 anlaş

-0.34

 laikā

-0.34

 Económica

-0.33

POSITIVE LOGITS

AddTagHelper

0.69

 autorytatywna

0.64

setVerticalGroup

0.64

RTGC

0.62

uxxxx

0.62

complexContent

0.61

 faſt

0.60

ftagPool

0.60

 Verſ

0.59

ſelves

0.58

Activations Density 0.083%

negative signs or expressions

non-

No Comments

No Known Activations