INDEX

Explanations

programming-related terms, specifically around function definitions and output values

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

returns or describes function output

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

Personensuche

-0.58

EDEFAULT

-0.50

 pinulongan

-0.47

 ProtoMessage

-0.47

CPtr

-0.47

 الرياضيه

-0.45

 AspNetCore

-0.45

fjspx

-0.45

 MongoClient

-0.45

 виправивши

-0.44

POSITIVE LOGITS

 kasarigan

0.43

 bogotá

0.40

httphttps

0.38

 neither

0.38

 hurtig

0.37

AnchorStyles

0.37

 atmosfer

0.36

㈦

0.36

 safe

0.35

 permanently

0.35

Activations Density 0.065%

programming-related terms, specifically around function definitions and output values

returns or describes function output

No Comments

No Known Activations