INDEX

Explanations

references to mountain ranges and geographical features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 مرئيه

-0.60

coroutines

-0.55

iprot

-0.53

 članak

-0.52

mosis

-0.52

giène

-0.49

 kasarigan

-0.49

ագրություններ

-0.48

GraphicsUnit

-0.48

+#+#

-0.48

POSITIVE LOGITS

 mountains

1.10

 mountain

1.05

 Mountains

1.03

mountains

1.03

 Mount

0.96

 Mountain

0.96

 mount

0.96

Mountains

0.94

 montagne

0.91

 MOUNT

0.91

Activations Density 0.443%

references to mountain ranges and geographical features

No Comments

No Known Activations