INDEX

Explanations

verbs in various forms, particularly those indicating states of being or action

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

]),

-0.42

 release

-0.41

ModelMap

-0.40

 Graham

-0.39

φ

-0.39

 fondo

-0.37

ственная

-0.36

 Scout

-0.36

 multiple

-0.35

re

-0.35

POSITIVE LOGITS

Datuak

1.15

StoryboardSegue

0.99

+#+#

0.94

AsUp

0.88

 houſe

0.87

Diwedd

0.85

Efq

0.84

 autorytatywna

0.83

 laſt

0.81

 ddelweddau

0.79

Activations Density 0.277%

verbs in various forms, particularly those indicating states of being or action

No Comments

No Known Activations