INDEX

Explanations

structural or formatting elements in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

Portály

-0.66

unknownFields

-0.57

 strand

-0.56

usepackage

-0.54

 atra

-0.50

ọc

-0.48

pri

-0.48

dic

-0.48

 functionality

-0.48

zig

-0.47

POSITIVE LOGITS

 оригіналу

0.78

 maux

0.76

 respectivement

0.74

otheby

0.73

 Shakspeare

0.72

 AssemblyCulture

0.71

vgili

0.69

 wherefore

0.69

 Athenians

0.68

 quelcon

0.67

Activations Density 0.565%

structural or formatting elements in the text

No Comments

No Known Activations