INDEX

Explanations

technical specifications related to design and functionality

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

after "end" or parts of objects

np_acts-logits-general · gemini-2.0-flash

end of portions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

Net

-0.42

Net

-0.41

clonal

-0.39

 menem

-0.39

 reconstru

-0.38

ombic

-0.38

throw

-0.38

 prosi

-0.37

 Booth

-0.36

ragona

-0.36

POSITIVE LOGITS

istoitu

0.58

 saites

0.57

findpost

0.51

Atsauces

0.51

 AssemblyCulture

0.48

NewGuid

0.46



0.46

styleType

0.45

⤹

0.44

ConstraintMaker

0.44

Activations Density 0.426%

technical specifications related to design and functionality

after "end" or parts of objects

end of portions

No Comments

No Known Activations