INDEX

Explanations

references to specific characters or entities from a narrative or text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

itects

-0.53

actylus

-0.51

fromCharCode

-0.48

otides

-0.48

Etern

-0.46

もちろん

-0.46

direkt

-0.46

 rodillas

-0.46

ニーカー

-0.45

 niebla

-0.45

POSITIVE LOGITS

Su

0.86

Su

0.83

su

0.75

SU

0.71

 Suzy

0.66

 Suri

0.65

 esterni

0.61

 Suarez

0.61

 Sutherland

0.59

SUA

0.59

Activations Density 0.025%

references to specific characters or entities from a narrative or text

No Comments

No Known Activations