INDEX

Explanations

keywords related to correctness and appropriateness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

限り

-0.54

ably

-0.53

 DÍA

-0.51

 المثال

-0.49

 gouver

-0.49

__":

-0.48

증

-0.48

請繼續往下閱讀

-0.47

__':

-0.46

dado

-0.46

POSITIVE LOGITS

alyptus

0.67

UrlResolution

0.66

SizeF

0.64

 circonstances

0.62

 sockaddr

0.61

 Italijanski

0.61

Билгалдахарш

0.60

LEncoder

0.59

medriver

0.59

DJANGO

0.58

Activations Density 0.468%

keywords related to correctness and appropriateness

No Comments

No Known Activations