INDEX

Explanations

terms related to technical processes and measurements

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

after or before

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

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

 primary

-0.39

 secondary

-0.38

ret

-0.37

 Secondary

-0.36

otan

-0.35

 inse

-0.35

vik

-0.35

mädchen

-0.34

 rabb

-0.34

 radioactivity

-0.34

POSITIVE LOGITS

 '\\;'

0.76

 ModelExpression

0.72

 CreateTagHelper

0.63

 <=",

0.62

 estekak

0.61

awtextra

0.58

astify

0.55

ConstraintMaker

0.55

 мәкал

0.54

 GenerationType

0.53

Activations Density 0.608%

terms related to technical processes and measurements

after or before

No Comments

No Known Activations