INDEX

Explanations

references to specific geographic locations and trails

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

horabuena

-0.50

這就是

-0.44

ريقيا

-0.42

AsyncResult

-0.40

Soorten

-0.40

 estekak

-0.38

CLUYE

-0.38

OfYear

-0.36

这就是

-0.35

LikeLiked

-0.35

POSITIVE LOGITS

 western

0.76

 eastern

0.76

 southern

0.69

 northern

0.63

western

0.60

eastern

0.59

 north

0.57

 northwestern

0.56

 city

0.55

mybatisplus

0.55

Activations Density 0.459%

references to specific geographic locations and trails

No Comments

No Known Activations