INDEX

Explanations

references to political figures and context related to their influence and social gatherings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.17.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

 storici

-0.53

-0.50

 fidélité

-0.48

もとも

-0.48

もともと

-0.46

 sejarah

-0.43

 caráter

-0.43

 propres

-0.43

 historii

-0.43

History

-0.42

POSITIVE LOGITS

 hottest

1.15

 trend

1.12

 trends

1.12

 trending

1.08

 trendy

1.06

trend

1.04

trends

1.02

 vogue

1.00

trending

0.99

 fashionable

0.98

Activations Density 0.231%

references to political figures and context related to their influence and social gatherings

No Comments

No Known Activations