INDEX

Explanations

geographic locations mentioned in relation to stories or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

CopyWith

-0.36

⎩

-0.35

Sydney

-0.34

 Sydney

-0.33

íncia

-0.33

Omar

-0.33

gom

-0.33

 Dirichlet

-0.32

gor

-0.32

 Melbourne

-0.32

POSITIVE LOGITS

 Ipswich

0.73

 Queensland

0.73

Queensland

0.68

 Brisbane

0.65

 Sunshine

0.64

 CreateTagHelper

0.64

QLD

0.62

Sunshine

0.58

Brisbane

0.57

 Cairns

0.57

Activations Density 0.115%

geographic locations mentioned in relation to stories or events

No Comments

No Known Activations