INDEX

Explanations

phrases and sentiments expressing good news or positive developments

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

news or positive updates

np_acts-logits-general · gemini-2.0-flash

good news

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

prä

-0.48

 springfox

-0.46

 écoulé

-0.43

]._

-0.43

fering

-0.43

 Effectiveness

-0.43

 Interactions

-0.43

rifice

-0.42

 geï

-0.42

 Effort

-0.41

POSITIVE LOGITS

 news

3.10

news

2.30

 News

2.13

News

2.08

 NEWS

2.05

 noticias

1.95

 noticia

1.91

NEWS

1.84

 nieuws

1.82

 berita

1.77

Activations Density 0.191%

phrases and sentiments expressing good news or positive developments

news or positive updates

good news

No Comments

No Known Activations