INDEX

Explanations

phrases related to justifications and explanations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

providedIn

-0.62

 Talis

-0.55

 Elis

-0.46

 ModelExpression

-0.44

principalTable

-0.44

uVar

-0.44

IERC

-0.43

 manageable

-0.43

writerow

-0.43

Portail

-0.42

POSITIVE LOGITS

 reasons

1.56

 reason

1.52

 Reasons

1.43

Reasons

1.41

why

1.40

reasons

1.31

Reason

1.27

 Gründe

1.26

 Reason

1.23

 REASONS

1.22

Activations Density 0.473%

phrases related to justifications and explanations

No Comments

No Known Activations