INDEX

Explanations

instances of actions and events occurring in a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

rrh

-0.43

])));

-0.43

DockStyle

-0.43

‘

-0.42

MEL

-0.41

melse

-0.40

 assurent

-0.40

الحياه

-0.39

=\{

-0.39

esía

-0.39

POSITIVE LOGITS

UnknownFields

0.78

Efq

0.75

 innocently

0.69

 AssemblyCulture

0.68

ſelf

0.68

IsMutable

0.66

 ་་

0.66

 myſelf

0.65

 Anſ

0.63

 doubtnut

0.63

Activations Density 0.299%

instances of actions and events occurring in a narrative context

No Comments

No Known Activations