INDEX

Explanations

programming language syntax and metadata formatting

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

encoding to user

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

RegressionTest

-0.69

 Photocase

-0.65

RenderAtEndOf

-0.65

 betweenstory

-0.63

 InputDecoration

-0.61

 surla

-0.61

Aholisi

-0.60

createServer

-0.58

webElementXpaths

-0.57

 myſelf

-0.57

POSITIVE LOGITS

 coding

0.64

-*-

0.48

 Coding

0.46

coding

0.45

Coding

0.43

 charset

0.40

 code

0.36

encoding

0.34

Code

0.32

GraphicsUnit

0.31

Activations Density 0.000%

programming language syntax and metadata formatting

encoding to user

No Comments

No Known Activations