INDEX

Explanations

specific names and titles in the context of a medical or scientific discussion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/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.36.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

-0.63

bootstrapcdn

-0.55

expandindo

-0.55

 Италијани

-0.51

ArrowToggle

-0.50

 تضيفلها

-0.49

 invokingState

-0.48

⤹

-0.48

 BorderSide

-0.47

MessageTagHelper

-0.47

POSITIVE LOGITS

watt

0.36

trained

0.35

Trained

0.32

 Deiner

0.32

 Trieb

0.32

Pom

0.32

Ord

0.31

찾

0.31

Msg

0.31

 Treffen

0.31

Activations Density 0.432%

specific names and titles in the context of a medical or scientific discussion

No Comments

No Known Activations