INDEX

Explanations

references to mountains or mountainous features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 DELHI

-0.79

躇

-0.68

новниш

-0.65

Życiorys

-0.64

 Gabor

-0.63

 CHICAGO

-0.63

 Hygge

-0.63

cleros

-0.62

 Middlesex

-0.62

COLN

-0.62

POSITIVE LOGITS

 mountain

2.05

 Mountain

2.04

 mountains

1.89

Mountain

1.88

mountain

1.72

 Mountains

1.68

 MOUNTAIN

1.67

mountains

1.53

Mountains

1.47

 mountainous

1.40

Activations Density 0.064%

references to mountains or mountainous features

No Comments

No Known Activations