INDEX

Explanations

content related to featured blog posts and their highlights

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

featured

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

Etymology

-0.52

ắn

-0.51

Решение

-0.50

SharedDtor

-0.49

帖最后由

-0.49

XmlAccessType

-0.49

 RSSSF

-0.49

Tembelea

-0.49

 gmina

-0.48

INSTRUCTIONS

-0.48

POSITIVE LOGITS

 featured

1.03

 prominently

0.89

 feature

0.84

featured

0.84

 Featured

0.83

feature

0.73

Featured

0.72

 Feature

0.72

 FEATURE

0.68

 features

0.68

Activations Density 0.009%

content related to featured blog posts and their highlights

featured

No Comments

No Known Activations