INDEX

Explanations

repeated use of the word "much"

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

 poichè

-0.64

 vettor

-0.58

Occasionally

-0.55

 italie

-0.55

 Banach

-0.54

 ajudá

-0.54

al

-0.53

ed

-0.53

 vettoriale

-0.52

harapkan

-0.52

POSITIVE LOGITS

Mucho

1.00

 المعيارى

1.00

 Amounts

0.97

 estimés

0.96

AMOUNT

0.96

 amount

0.95

ánto

0.91

 AMOUNT

0.90

principalTable

0.89

amounts

0.89

Activations Density 0.102%

repeated use of the word "much"

No Comments

No Known Activations