INDEX

Explanations

specific identifiers and terms related to technical or programming aspects

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

mathematical and programming symbols

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

:✨

-0.64

 AssemblyTitle

-0.58

Życiorys

-0.56

 समीक्षाओं

-0.54

 reassurance

-0.54

Brainz

-0.52

ьаж

-0.51

 ब्रेकडाउन

-0.50

devamını

-0.50

 criti

-0.50

POSITIVE LOGITS

EndContext

0.46

styleType

0.45

 betweenstory

0.42

FieldDescriptor

0.34

rungsseite

0.34

RTHOOK

0.33

 değildir

0.33

 Chwiliwch

0.33

 ivelany

0.33

 varandra

0.32

Activations Density 0.038%

specific identifiers and terms related to technical or programming aspects

mathematical and programming symbols

No Comments

No Known Activations