INDEX

Explanations

words or phrases indicating numerical thresholds or limits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_14/width_16k/average_l0_67

Prompts (Dashboard)

24,576 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

QList

-0.31

อะ

-0.30

ñera

-0.29

듈

-0.29

期刊论文

-0.29

 Anfitrión

-0.28

 host

-0.27

 beng

-0.27

很赞哦

-0.27

 foule

-0.27

POSITIVE LOGITS

 until

0.76

 till

0.75

 até

0.73

 jusqu

0.73

 untill

0.72

 hasta

0.71

until

0.71

Até

0.70

jusqu

0.70

 Jusqu

0.69

Activations Density 0.300%

words or phrases indicating numerical thresholds or limits

No Comments

No Known Activations