INDEX

Explanations

phrases indicating relationships or comparisons among different categories or items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

GIVEREF

-0.68

 endwhile

-0.68

клопе

-0.64

сылкі

-0.64

riwal

-0.62

 فريبيس

-0.61

postmedia

-0.60

ğlık

-0.60

 couteau

-0.59

sterious

-0.58

POSITIVE LOGITS

 közül

0.65

ParallelGroup

0.59

うち

0.56

中でも

0.50

imageshack

0.50

の中でも

0.50

のうち

0.48

__).

0.48

の中で

0.48

 незавершена

0.47

Activations Density 0.431%

phrases indicating relationships or comparisons among different categories or items

No Comments

No Known Activations