INDEX

Explanations

phrases related to the granting of permissions or legal approvals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

httphttps

-0.43

 perſ

-0.42

 filtered

-0.41

 encre

-0.40

İstinadlar

-0.40

KOS

-0.40

fum

-0.40

 totem

-0.39

 pops

-0.39

 Gond

-0.38

POSITIVE LOGITS

typical

0.73

 granted

0.70

 grant

0.68

granted

0.68

 Typical

0.64

 typical

0.63

 typique

0.63

 grants

0.63

Typical

0.61

grants

0.57

Activations Density 0.188%

phrases related to the granting of permissions or legal approvals

No Comments

No Known Activations