INDEX

Explanations

terms related to cancer and specifically adenocarcinoma

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

Slf

-0.62

vœ

-0.52

 protoimpl

-0.52

Pvt

-0.45

Pty

-0.44

 piety

-0.44

 laid

-0.43

 ujednoznacz

-0.42

 ethic

-0.42

 fédé

-0.41

POSITIVE LOGITS

 Zimmermann

1.03

 McCullough

0.97

————————————————

0.96

AAAAAAAAAAAAAAAA

0.95

 McDonough

0.94

 Cunningham

0.94

 Schwarzenegger

0.94

…………………………………………

0.90

 backgroundColor

0.90

 Bartholomew

0.90

Activations Density 2.644%

terms related to cancer and specifically adenocarcinoma

No Comments

No Known Activations