INDEX

Explanations

document sections that begin with a specific token or structure, indicating the start of new content or sections

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

 AssemblyCulture

-0.69

tagHelperRunner

-0.68

RenderAtEndOf

-0.68

 InputDecoration

-0.57

mybatisplus

-0.57

AndEndTag

-0.55

AddTagHelper

-0.53

httphttps

-0.51

writeFieldEnd

-0.51

تقاوى

-0.44

POSITIVE LOGITS

 snippetHide

0.52

PYX

0.51

ंदीखरीदारी

0.51

 humaines

0.50

 énergé

0.49

 miniaturka

0.47

 humanidad

0.45

TintMode

0.45

 occuper

0.45

 cardiaque

0.44

Activations Density 1.007%

document sections that begin with a specific token or structure, indicating the start of new content or sections

No Comments

No Known Activations