INDEX

Explanations

specific technical terms and jargon, particularly related to programming and data structures

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 numberOfRows

-0.35

InSection

-0.35

OptionsMenu

-0.33

Sơ

-0.33

 floor

-0.33

wallpapers

-0.32

 đà

-0.32

 cast

-0.31

圖

-0.31

upaten

-0.31

POSITIVE LOGITS

LookAnd

0.63

aarrggbb

0.62

 فريبيس

0.53

 nakalista

0.52

 autorytatywna

0.48

hyrchwyd

0.47

providedIn

0.46

 ExecuteAsync

0.45

 AssemblyProduct

0.45

webElementGuid

0.45

Activations Density 0.605%

specific technical terms and jargon, particularly related to programming and data structures

No Comments

No Known Activations