INDEX

Explanations

programming and mathematical function notations

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

expandindo

-0.59

 OnDestroy

-0.56

ötzlich

-0.54

 kasarigan

-0.53

nicole

-0.53

Nicky

-0.52

ThemeOverlay

-0.52

 queſta

-0.50

cheinend

-0.49

theless

-0.49

POSITIVE LOGITS

0.88

0.86

0.84

 getF

0.79

0.72

फ

0.66

FE

0.65

getF

0.64

ฟ

0.62

Ф

0.61

Activations Density 2.650%

programming and mathematical function notations

No Comments

No Known Activations