INDEX

Explanations

references to structured data or programming concepts

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

HTML tags and code snippets

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

 Taktlose

-0.87

featureID

-0.84

MigrationBuilder

-0.84

 Chwiliwch

-0.84

IsMutable

-0.81

httphttps

-0.78

Hentet

-0.71

 OMITBAD

-0.71

 للاسماء

-0.71

AndEndTag

-0.70

POSITIVE LOGITS

0.39

 esprits

0.36

 consisting

0.35

his

0.35

see

0.35

block

0.35

he

0.34

 départements

0.34

for

0.34

 inside

0.34

Activations Density 0.068%

references to structured data or programming concepts

HTML tags and code snippets

No Comments

No Known Activations