INDEX

Explanations

technical terms and abbreviations related to data units and communication protocols

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

followed by suffixes like -ment, -ic, -ing, -ical, -ate, -hip, -ory, -est

np_acts-logits-general · gemini-2.0-flash

disillusionment, prohibiting, astronomically, overlordship, Inhibitory, narrowest, flocking, prognostication

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

 Shih

-0.47

 Ando

-0.47

omości

-0.44

 Pats

-0.43

 createdBy

-0.43

えず

-0.42

変わらず

-0.42

เ

-0.42

öbb

-0.41

 メンズ

-0.41

POSITIVE LOGITS

ſelf

0.82

theless

0.77

 kapture

0.71

ing

0.69

addContainerGap

0.66

 ſich

0.66

 Verſ

0.66

 dezelve

0.64

izing

0.64

retudo

0.63

Activations Density 2.072%

technical terms and abbreviations related to data units and communication protocols

followed by suffixes like -ment, -ic, -ing, -ical, -ate, -hip, -ory, -est

disillusionment, prohibiting, astronomically, overlordship, Inhibitory, narrowest, flocking, prognostication

No Comments

No Known Activations