INDEX

Explanations

references to timeframes, particularly midpoints within months or years

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.27.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

elemField

-0.47

 lệ

-0.41

验证码

-0.40

إنه

-0.38

 AssemblyCulture

-0.36

Leistung

-0.35

machung

-0.35

 MenuView

-0.33

 Fuer

-0.33

liferay

-0.32

POSITIVE LOGITS

mid

0.78

 ویکی‌پدی

0.68

mid

0.65

Mid

0.64

Mid

0.63

PYX

0.57

MID

0.55

sizeCache

0.55

 midt

0.54

MID

0.53

Activations Density 0.083%

references to timeframes, particularly midpoints within months or years

No Comments

No Known Activations