INDEX

Explanations

references to specific topics or items previously mentioned in the text

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

KommentareTeilen

-0.46

ArrowToggle

-0.39

 those

-0.38

 next

-0.38

એ

-0.37

next

-0.37

voet

-0.36

 nächsten

-0.36

:✨

-0.36

 těch

-0.36

POSITIVE LOGITS

 kind

0.83

 regard

0.68

kind

0.57

 circonst

0.57

way

0.57

 właśnie

0.56

AutoScaleMode

0.54

 matter

0.52

 regards

0.52

 ModelExpression

0.52

Activations Density 0.264%

references to specific topics or items previously mentioned in the text

No Comments

No Known Activations