INDEX

Explanations

quantitative aspects and comparisons in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 nào

-0.57

newswire

-0.57

র্ব

-0.56

 العنوان

-0.56

stalt

-0.55

awaiter

-0.55

Slightly

-0.54

 Reverso

-0.52

پان

-0.52

거

-0.52

POSITIVE LOGITS

 many

0.95

 MANY

0.92

MANY

0.90

many

0.85

 Many

0.83

 Viele

0.81

 vieler

0.79

 Banyak

0.78

 viele

0.78

Many

0.78

Activations Density 0.949%

quantitative aspects and comparisons in various contexts

No Comments

No Known Activations