INDEX

Explanations

proper nouns and specific references related to locations, entities, and names

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

mathematical notation or names

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

 CreateTagHelper

-0.85

TagMode

-0.79

enderror

-0.76

 وتسجيلات

-0.75

 himo

-0.72

+#+

-0.71

IsContent

-0.71

EndInit

-0.67

 فريبيس

-0.66

aarrggbb

-0.65

POSITIVE LOGITS

 ensues

0.28

↵↵

0.28

Weiterlesen

0.28

据

0.27

Hon

0.26

Redacción

0.26

成

0.25

Witam

0.25

ret

0.24

)}+\

0.24

Activations Density 0.129%

proper nouns and specific references related to locations, entities, and names

mathematical notation or names

No Comments

No Known Activations