INDEX

Explanations

instances of structured data or coding related to programming and configurations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

ба

-0.50

-0.49

-0.48

by

-0.47

ha

-0.47

nak

-0.46

sk

-0.45

-0.44

ll

-0.44

POSITIVE LOGITS

 itſelf

1.17

ſelves

1.12

ſelf

1.10

 myſelf

1.03

 pleaſure

1.02

 purpoſe

1.00

 ſmall

0.97

BibitemShut

0.97

Efq

0.96

脚注の使い方

0.96

Activations Density 0.003%

instances of structured data or coding related to programming and configurations

No Comments

No Known Activations