INDEX

Explanations

phrases indicating intention or future actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

ivably

-0.50

agerie

-0.49

⌐

-0.49

buri

-0.46

 Alar

-0.43

Moy

-0.43

Ueb

-0.43

го

-0.43

 кла

-0.42

bal

-0.42

POSITIVE LOGITS

sizeCache

0.81

principalTable

0.81

parsedMessage

0.79

WithIOException

0.75

uxxxx

0.72

featureID

0.72

 AssemblyCompany

0.71

FieldBuilder

0.71

makeText

0.70

 beginnetje

0.70

Activations Density 0.038%

phrases indicating intention or future actions

No Comments

No Known Activations