INDEX

Explanations

content related to instructional guides and helpful tips

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

IsMutable

-0.49

 <>",

-0.43

RegressionTest

-0.40

offsetof

-0.37

 Biôgrafia

-0.36

хьтан

-0.34

omination

-0.34

 impeach

-0.34

writeFieldEnd

-0.33

 Taktlose

-0.32

POSITIVE LOGITS

 advice

1.23

 tips

1.14

advice

1.11

 guides

1.10

 guidance

1.09

 guide

1.09

 instructions

1.05

 advices

1.03

 consejos

1.03

 ADVICE

1.01

Activations Density 0.510%

content related to instructional guides and helpful tips

No Comments

No Known Activations