INDEX

Explanations

numerical data, particularly in the context of statistics and analysis

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

citations and journal abbreviations

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

󠁢

-0.71

DockStyle

-0.60

 Chwiliwch

-0.59

inSlope

-0.57

yarnpkg

-0.56

 فريبيس

-0.54

 كومونز

-0.54

 uParam

-0.54

GEBURTS

-0.52

AddTagHelper

-0.52

POSITIVE LOGITS

ScopeManager

0.52

 CURIAM

0.41

 disambiguazione

0.40

webElementXpaths

0.37

Brief

0.37

Citation

0.35

 keli

0.35

Short

0.33

 korte

0.33

ABUL

0.33

Activations Density 0.075%

numerical data, particularly in the context of statistics and analysis

citations and journal abbreviations

No Comments

No Known Activations