INDEX

Explanations

punctuation marks, particularly periods

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

filed under

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

bootstrapcdn

-0.44

 dictadura

-0.35

tilha

-0.35

Gweler

-0.34

acyj

-0.34

 Neri

-0.32

Jährige

-0.32

 قلم

-0.31

 Kral

-0.31

Demografía

-0.31

POSITIVE LOGITS

 autorytatywna

1.17

RegressionTest

0.96

GEBURTSDATUM

0.93

NameInMap

0.88

 Photocase

0.85

httphttps

0.83

:✨

0.82

 EconPapers

0.80

 AssemblyTitle

0.76

AsUp

0.75

Activations Density 0.001%

punctuation marks, particularly periods

filed under

No Comments

No Known Activations