INDEX

Explanations

terms related to categorization and classification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 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

vičius

-0.50

éramos

-0.46

 McIn

-0.44

 Zoll

-0.42

 tids

-0.41

AddWithValue

-0.41

 cheval

-0.40

IsContent

-0.39

elum

-0.39

ását

-0.39

POSITIVE LOGITS

ies

1.25

IES

1.12

ys

1.07

ying

0.99

0.98

yship

0.94

ies

0.91

ries

0.87

YS

0.87

yp

0.84

Activations Density 0.412%

terms related to categorization and classification

No Comments

No Known Activations