INDEX

Explanations

statements that express uncertainty or inquiry regarding a situation or topic

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 autorytatywna

-0.90

 للاسماء

-0.90

Vidite

-0.74

 pleaſure

-0.74

存于互联网档案馆

-0.71

<unused52>

-0.70

[@BOS@]

-0.69

<unused68>

-0.69

<unused51>

-0.69

<unused28>

-0.69

POSITIVE LOGITS

 maybe

0.52

maybe

0.49

perhaps

0.48

 perhaps

0.47

 possibly

0.47

or

0.46

 Maybe

0.46

 might

0.45

Maybe

0.43

 Possibly

0.41

Activations Density 0.567%

statements that express uncertainty or inquiry regarding a situation or topic

No Comments

No Known Activations