INDEX

Explanations

technical terms and specific details related to scientific research and findings

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

jface

-0.39

SuspendLayout

-0.39

agon

-0.39

 rerum

-0.38

 pinulongan

-0.38

AddTagHelper

-0.37

pected

-0.35

 forgotten

-0.35

:✨

-0.35

spoken

-0.33

POSITIVE LOGITS

led

1.80

 helped

1.79

 prompted

1.61

 enabled

1.59

helped

1.57

 contributed

1.48

 Helped

1.43

 caused

1.41

 influenced

1.38

 spurred

1.34

Activations Density 1.246%

technical terms and specific details related to scientific research and findings

No Comments

No Known Activations