INDEX

Explanations

specific articles like "a," "an," and "the."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 myſelf

-1.39

 pleaſure

-1.37

 himſelf

-1.35

 purpoſe

-1.34

 itſelf

-1.29

 Jefus

-1.29

 Theſe

-1.29

 Monfieur

-1.28

 faſt

-1.27

 iſt

-1.23

POSITIVE LOGITS

0.65

in

0.63

0.56

0.55

0.52

to

0.51

as

0.51

↵

0.50

 with

0.49

0.47

Activations Density 0.068%

specific articles like "a," "an," and "the."

No Comments

No Known Activations