INDEX

Explanations

references to general concepts and items

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

"thing" or "things."

np_acts-logits-general · gemini-2.0-flash

introduces 'thing' or 'things'

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

 gales

-0.85

 pleaſure

-0.85

adecimal

-0.80

 ſtate

-0.80

 myſelf

-0.79

 ―――――

-0.79

 ſta

-0.78

 uſe

-0.78

 juſ

-0.78

 preſ

-0.77

POSITIVE LOGITS

 things

1.45

 thing

1.43

 Thing

1.40

Things

1.36

 Things

1.35

 THING

1.32

 THINGS

1.31

Thing

1.23

things

1.17

THING

1.06

Activations Density 0.058%

references to general concepts and items

"thing" or "things."

introduces 'thing' or 'things'

No Comments

No Known Activations