INDEX

Explanations

occurrences of the word "two" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

ly

-0.76

Various

-0.45

 Various

-0.43

various

-0.40

ten

-0.40

 total

-0.38

 various

-0.38

mes

-0.37

place

-0.37

mer

-0.37

POSITIVE LOGITS

 halves

1.03

 thirds

1.01

thirds

0.94

 extremes

0.88

 hemispheres

0.78

 opposite

0.73

 opposing

0.73

 contrasting

0.72

 sides

0.71

 opposites

0.70

Activations Density 0.327%

occurrences of the word "two" in various contexts

No Comments

No Known Activations