INDEX

Explanations

instances of layout-related terminology

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

wicz

-0.72

 McLeod

-0.55

 amarelo

-0.55

 کیلو

-0.55

 lampada

-0.55

-0.54

+#+

-0.53

 sher

-0.53

 gomma

-0.53

bės

-0.53

POSITIVE LOGITS

 Layout

3.02

layout

3.01

 layout

2.98

Layout

2.97

 layouts

2.76

 LAYOUT

2.56

LAYOUT

2.37

layouts

2.08

Layouts

2.03

ayout

1.89

Activations Density 0.052%

instances of layout-related terminology

No Comments

No Known Activations