INDEX

Explanations

names and proper nouns

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

Squirrel

-0.77

 Melville

-0.76

 Barnet

-0.74

mybatisplus

-0.73

 septum

-0.72

 Kela

-0.72

 Marta

-0.72

 Barrington

-0.72

 Packard

-0.72

 ddelweddau

-0.71

POSITIVE LOGITS

Ele

0.73

كويكب

0.61

Chal

0.60

 charge

0.59

 honours

0.58

 Charge

0.57

BoxShadow

0.57

embar

0.57

bicara

0.57

Pras

0.56

Activations Density 1.808%

names and proper nouns

No Comments

No Known Activations