INDEX

Explanations

underscore characters or their variations within text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 aikaa

-0.58

 yaşında

-0.58

antaranya

-0.56

tubo

-0.56

 τὴν

-0.55

VERSION

-0.54

 Commonly

-0.53

 longue

-0.53

","+

-0.53

ENEFITS

-0.52

POSITIVE LOGITS

Personendaten

0.92

 muualla

0.76

تقاوى

0.74

/*

0.71

 nakalista

0.67

 المعيارى

0.67

 intptr

0.66

 really

0.66

IVEREF

0.65

 AssemblyCompany

0.65

Activations Density 0.221%

underscore characters or their variations within text

No Comments

No Known Activations