INDEX

Explanations

quantifiers and negations in the text

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

but, not, be, they

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 صوتيه

-0.81

 فريبيس

-0.76

المكان

-0.72

Diweddarwch

-0.72

تفصیلات

-0.69

LOYEES

-0.66

 مشين

-0.66

 snippetHide

-0.61

TagMode

-0.61

GEBURTSDATUM

-0.60

POSITIVE LOGITS

ówczas

0.40

 zaś

0.35

 MessageBoxIcon

0.34

 geslacht

0.30

áctenos

0.29

 eletrônico

0.29

 venezolano

0.29

 procedentes

0.29

łaszcza

0.29

期刊论文

0.29

Activations Density 0.137%

quantifiers and negations in the text

but, not, be, they

No Known Activations