INDEX

Explanations

HTML tags or markup language syntax

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 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

 antidesliz

-0.47

 themſelves

-0.45

 partitions

-0.45

 faſt

-0.44

ſelf

-0.43

 ſever

-0.43

LookAnd

-0.42

 convid

-0.42

 telefonu

-0.42

 المعيارى

-0.42

POSITIVE LOGITS

hr

1.36

HR

0.99

HR

0.98

hr

0.96

Hr

0.85

Hr

0.79

HRP

0.50

بوابة

0.50

Bakgrunnsstoff

0.50

 tartalomajánló

0.47

Activations Density 0.119%

HTML tags or markup language syntax

No Comments

No Known Activations