INDEX

Explanations

phrases indicating caution or limitation in the context of scientific or legal discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

IUrlHelper

-1.03

RenderAtEndOf

-0.92

Autoritní

-0.88

الحياه

-0.83

AccessorTable

-0.81

+#+#

-0.77

Diweddarwch

-0.77

OGND

-0.77

 snippetHide

-0.76

 ComVisible

-0.74

POSITIVE LOGITS

0.40

<eos>

0.33

0.31

0.30

ha

0.29

ẾT

0.29

con

0.29

<sub>

0.28

 chance

0.27

Activations Density 1.151%

phrases indicating caution or limitation in the context of scientific or legal discussions

No Comments

No Known Activations