INDEX

Explanations

references to specific seasons in a series or event

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

the word "season"

np_acts-logits-general · gemini-2.0-flash

season

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

 white

-0.49

utuhkan

-0.48

 Públicas

-0.47

 White

-0.47

Gib

-0.46

 Franke

-0.45

big

-0.45

 lápiz

-0.45

white

-0.43

 Middle

-0.42

POSITIVE LOGITS

 season

1.28

Season

1.26

season

1.25

 SEASON

1.25

 Season

1.24

SEASON

1.14

seasons

1.14

 seasons

1.12

 Seasons

1.09

Seasons

1.09

Activations Density 0.038%

references to specific seasons in a series or event

the word "season"

season

No Comments

No Known Activations