INDEX

Explanations

references to reading and literature

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

reading books

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

parsedMessage

-0.71

 estekak

-0.71

 ivelany

-0.61

Искәрмәләр

-0.61

IntoConstraints

-0.60

 argint

-0.60

VYMaps

-0.59

 bezeichneter

-0.58

ulant

-0.57

TargetException

-0.57

POSITIVE LOGITS

 reading

0.66

 books

0.65

 literary

0.54

Reading

0.52

 Reading

0.51

 book

0.49

 Bücher

0.47

 lectura

0.46

 reader

0.46

books

0.46

Activations Density 0.016%

references to reading and literature

reading books

No Comments

No Known Activations