INDEX

Explanations

legal terms and phrases indicating criminal activity or formal proceedings

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

Followed by prepositions or the word "the"

np_acts-logits-general · gemini-2.0-flash

advice or recommendation

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

 disambiguazione

-0.72

 resourceCulture

-0.52

RegressionTest

-0.52

FailureListener

-0.50

AddTagHelper

-0.49

GetAxis

-0.49

setVerticalGroup

-0.48

StoryboardSegue

-0.47

respec

-0.46

enterOuterAlt

-0.45

POSITIVE LOGITS

 recommendation

1.47

 recommendations

1.30

 recommended

1.28

 suggestion

1.24

 advice

1.22

 suggested

1.16

 advised

1.14

 suggestions

1.12

recommendation

1.09

 recomendación

1.08

Activations Density 0.471%

legal terms and phrases indicating criminal activity or formal proceedings

Followed by prepositions or the word "the"

advice or recommendation

No Comments

No Known Activations