INDEX

Explanations

references to specific locations and political figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

cove

-0.45

 cuchar

-0.44

pot

-0.43

portál

-0.42

stav

-0.41

oregon

-0.40

 tarko

-0.40

Bangalore

-0.40

 bumbu

-0.40

POSITIVE LOGITS

lamabad

1.22

########.

1.17

 Pakistan

1.09

awtextra

1.07

 Pakistani

1.04

Pakistan

1.03



1.02

 Islamabad

0.98

 Lahore

0.97

 pakistan

0.94

Activations Density 0.103%

references to specific locations and political figures

No Comments

No Known Activations