INDEX

Explanations

references to academic or professional titles and roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/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.15.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

 emberek

-0.50

 Gouvernement

-0.48

 animale

-0.48

 kteří

-0.45

excess

-0.43

 excess

-0.42

 Tories

-0.41

 fools

-0.41

GPs

-0.41

 ludzi

-0.40

POSITIVE LOGITS

 former

0.94

InjectAttribute

0.88

 fellow

0.82

 Former

0.82

former

0.81

Former

0.81

 bassist

0.76

 Assistant

0.76

 longtime

0.75

webElementXpaths

0.74

Activations Density 0.214%

references to academic or professional titles and roles

No Comments

No Known Activations