INDEX

Explanations

the word "once" and variations of "you" to indicate repeated scenarios or occurrences

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

Text followed by "you"

np_acts-logits-general · gemini-2.0-flash

the word 'once'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 resultMap

-0.40

 jPanel

-0.40

TestResult

-0.37

 IOError

-0.35

FormField

-0.34

inflater

-0.33

 getB

-0.32

mActivity

-0.32

 PHPUnit

-0.32

fprintf

-0.30

POSITIVE LOGITS

Once

1.14

 Once

1.12

 ONCE

1.10

once

1.07

ONCE

1.05

 once

1.01

 Sobald

0.77

EndInit

0.75

 Upon

0.72

 sobald

0.71

Activations Density 0.072%

the word "once" and variations of "you" to indicate repeated scenarios or occurrences

Text followed by "you"

the word 'once'

No Comments

No Known Activations