INDEX

Explanations

questions and queries related to acquiring information or assistance

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

 propOrder

-0.58

MessageTagHelper

-0.55

UrlResolution

-0.55

SourceChecksum

-0.48

 kasarigan

-0.45

amitié

-0.43

astify

-0.43

новниш

-0.43

 MenuView

-0.42

 noemen

-0.42

POSITIVE LOGITS

NameInMap

0.48

HandlerContext

0.46

AttributeSet

0.45

 pamphlets

0.43



0.43

 Khor

0.43

 AttributeSet

0.42

 estekak

0.42

hasMoreElements

0.41

jsonwebtoken

0.40

Activations Density 0.610%

questions and queries related to acquiring information or assistance

No Comments

No Known Activations