INDEX

Explanations

instances of commas, suggesting it looks for textual structures or lists

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

PerformLayout

-0.58

InputBorder

-0.54

ChildScrollView

-0.51

 مشين

-0.50

 <<<<<<<<<<<<<<

-0.50

httphttps

-0.48

 Coating

-0.44

 coating

-0.41

PropertyChanging

-0.40

createStatement

-0.39

POSITIVE LOGITS

########.

0.54

 embarazadas

0.47

 editors

0.47

 readers

0.45

 Meksika

0.44

 Portuguesa

0.44

 assistants

0.44

 audiences

0.43

 însă

0.43

anément

0.43

Activations Density 0.279%

instances of commas, suggesting it looks for textual structures or lists

No Comments

No Known Activations