INDEX

Explanations

elements and functions related to image processing in a programming context

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

bolt_strike, SYMBOLS_FONT_NAME

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

Negative Logits

 snippetHide

-0.67

 disambiguazione

-0.60

GEBURTSDATUM

-0.59

 MainAxisSize

-0.57

########.

-0.57

MigrationBuilder

-0.56

 ModelExpression

-0.55

createStatement

-0.55

migrationBuilder

-0.54

 Roskov

-0.54

POSITIVE LOGITS

 worn

0.47

 astore

0.40

©

0.39

Our

0.39

On

0.39

fjspx

0.38

 lunettes

0.38

 reused

0.37

hütte

0.37

Li

0.37

Activations Density 0.004%

elements and functions related to image processing in a programming context

bolt_strike, SYMBOLS_FONT_NAME

No Known Activations