INDEX

Explanations

terms and phrases indicating validity or invalidity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 tour

-0.43

 them

-0.43

people

-0.41

 about

-0.41

how

-0.39

ChatMessage

-0.39

(!_

-0.38

 Bier

-0.38

Tour

-0.37

POSITIVE LOGITS

DataAnnotations

1.05

 Valid

1.04

 Validate

0.96

Valid

0.95

 VALID

0.94

 isValid

0.93

isValid

0.90

 Validator

0.89

valid

0.89

 Validity

0.89

Activations Density 0.158%

terms and phrases indicating validity or invalidity

No Comments

No Known Activations