INDEX

Explanations

details related to evidence, particularly that which involves illegal substances or weapons

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

following "containing", "were", "of", or "with"

np_acts-logits-general · gemini-2.0-flash

collections of items

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

forests

-0.76

 AssemblyTitle

-0.75

 defaultstate

-0.73

+#+#

-0.73

 فريبيس

-0.72

 للمعارف

-0.70

isissez

-0.67

contentLoaded

-0.65

 surla

-0.65

 indisponible

-0.64

POSITIVE LOGITS

 miscellaneous

0.64

 assorted

0.61

 containing

0.55

box

0.53

 unopened

0.51

old

0.51

 various

0.48

 items

0.47

InstrumentedTest

0.47

 spare

0.46

Activations Density 0.255%

details related to evidence, particularly that which involves illegal substances or weapons

following "containing", "were", "of", or "with"

collections of items

No Comments

No Known Activations