INDEX

Explanations

programming-related keywords and definitions

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

class or 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

 المعيارى

-0.89

 betweenstory

-0.88

rungsseite

-0.77

 queſta

-0.77

TypedDataSet

-0.77

transQ

-0.76

 gynhyrchwyd

-0.75

ſelben

-0.72

ſammen

-0.72

 beſti

-0.71

POSITIVE LOGITS

class

0.27

 class

0.25

rę

0.22

ウル

0.22

エー

0.22

 Class

0.22

 Pues

0.22

}}],

0.20

 efectivamente

0.20

setEditable

0.20

Activations Density 0.001%

programming-related keywords and definitions

class or code

No Comments

No Known Activations