INDEX

Explanations

syntax or structure related to programming and code execution

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

mathematical expressions and programming code

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

 defaultstate

-0.32

 Voulez

-0.29

++]=

-0.28

 ście

-0.28

rimo

-0.28

etk

-0.28

PositiveButton

-0.27

 soort

-0.27

("")){

-0.27

niei

-0.27

POSITIVE LOGITS

 propOrder

0.92

 juſ

0.68

 itſelf

0.62

 faſt

0.61

 ſch

0.60

 ſta

0.59

 poffible

0.59

 myſelf

0.58

Jereo

0.58

 ModelExpression

0.58

Activations Density 0.026%

syntax or structure related to programming and code execution

mathematical expressions and programming code

No Comments

No Known Activations