INDEX

Explanations

code-related elements and syntax within programming or markup files

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

misspellings

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

complexContent

-0.43

Pener

-0.38

 constitucional

-0.36

 pracow

-0.36

 contribuir

-0.35

 réserver

-0.35

 contribuer

-0.35

 uitge

-0.35

 classique

-0.35

 wyda

-0.35

POSITIVE LOGITS

دانشنامهٔ

0.78

WithIOException

0.75

TintMode

0.62

 betweenstory

0.60

 AssemblyCompany

0.59

PerformLayout

0.58

IntoConstraints

0.58



0.53

 Мексичка

0.53

AISSEE

0.52

Activations Density 0.006%

code-related elements and syntax within programming or markup files

misspellings

No Comments

No Known Activations