INDEX

Explanations

expressions of agreement and appreciation in dialogue

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

Personensuche

-0.66

 esternos

-0.52

 Administrativna

-0.51

OGND

-0.50

出版年

-0.49

 GenerationType

-0.48

 Numerade

-0.48

webElementXpaths

-0.47

SourceChecksum

-0.47

RepeatedField

-0.46

POSITIVE LOGITS

 argument

0.60

 arguments

0.55

 debate

0.52

 argue

0.50

 argued

0.48

 dilemma

0.47

 debated

0.46

 argues

0.45

romyalgia

0.44

arg

0.44

Activations Density 0.594%

expressions of agreement and appreciation in dialogue

No Comments

No Known Activations