INDEX

Explanations

keywords related to technical jargon, specifically in computing and genetics

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

Initialisms and acronyms

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

 Erbe

-0.59

∎

-0.57

fras

-0.56

usik

-0.50

IRUS

-0.50

 Werth

-0.49

白天

-0.48

TokenNameLBRACE

-0.48

 Kruse

-0.48

POSITIVE LOGITS

AddTagHelper

0.51

flashdata

0.49

CDCl

0.49

 unſ

0.47

wapV

0.46

 aDecoder

0.43

renewcommand

0.42

copg

0.42

WriteTagHelper

0.42

 Ceramby

0.41

Activations Density 0.457%

keywords related to technical jargon, specifically in computing and genetics

Initialisms and acronyms

No Comments

No Known Activations