INDEX

Explanations

references to information availability and offers access to detailed content or resources

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

Followed by "here", "now", "from", or "of"

np_acts-logits-general · gemini-2.0-flash

words following "effort" or "review"

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 propOrder

-1.09

PyExc

-0.98

featureID

-0.97

ंदीखरीदारी

-0.91

énario

-0.86

########.

-0.84

NameInMap

-0.83

 незавершена

-0.83

]")]

-0.82

 '\\;'

-0.82

POSITIVE LOGITS

 here

1.88

 HERE

1.86

here

1.52

HERE

1.50

 Here

1.37

 aquí

1.26

Here

1.21

 здесь

1.19

 below

1.19

 هنا

1.18

Activations Density 0.268%

references to information availability and offers access to detailed content or resources

Followed by "here", "now", "from", or "of"

words following "effort" or "review"

No Comments

No Known Activations

references to information availability and offers access to detailed content or resources

Followed by "here", "now", "from", or "of"

words following "effort" or "review"

No Comments

No Known Activations