INDEX

Explanations

references to 'line' and related concepts in a structured format

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

Datuak

-1.34

 Monfieur

-1.26

 juſt

-1.07

 myſelf

-1.06

bibinfo

-1.05

 himſelf

-1.04

edicated

-1.04

 pleaſure

-1.03

 preſent

-1.00

 leſs

-0.99

POSITIVE LOGITS

 Line

1.81

 LINE

1.78

 line

1.77

 lines

1.71

Line

1.67

line

1.63

 Lines

1.55

LINE

1.44

Lines

1.41

lines

1.39

Activations Density 0.094%

references to 'line' and related concepts in a structured format

No Comments

No Known Activations