INDEX

Explanations

phrases indicating clarity or significance in definitions and arguments

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

"In a _" or "_ ways" phrases

np_acts-logits-general · gemini-2.0-flash

in a sense / way / respects

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

 '\\;'

-0.86

 fometimes

-0.80

 ſeveral

-0.79

 poffible

-0.79

 pleaſure

-0.77

 myſelf

-0.74

 kasarigan

-0.74

 uſed

-0.73

 AssemblyTitle

-0.73

 purpoſe

-0.72

POSITIVE LOGITS

it

0.73

we

0.55

 this

0.53

而言

0.52

这是一

0.52

It

0.51

来说

0.50

來說

0.49

 genommen

0.49

0.48

Activations Density 0.213%

phrases indicating clarity or significance in definitions and arguments

"In a _" or "_ ways" phrases

in a sense / way / respects

No Comments

No Known Activations