INDEX

Explanations

references to time or position within text

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

"On" followed by a preposition/article

np_acts-logits-general · gemini-2.0-flash

on the one hand

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

новништво

-0.73

Jereo

-0.73

 itſelf

-0.72

ніципалі

-0.71

 NSCoder

-0.70

 kaarangay

-0.70

 Anſ

-0.68

 архивлан

-0.67

 ſeveral

-0.67

orsese

-0.66

POSITIVE LOGITS

 behalf

0.49

ollett

0.48

tero

0.44

 Pacific

0.44

top

0.44

<eos>

0.42

sup

0.41

 rigu

0.41

 dage

0.41

dde

0.41

Activations Density 0.125%

references to time or position within text

"On" followed by a preposition/article

on the one hand

No Comments

No Known Activations