INDEX

Explanations

garbled or non-alphanumeric characters

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

meeting/gathering specific groups

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

transQ

-0.91

aarrggbb

-0.88

 noDo

-0.86

UserScript

-0.79

 незавершена

-0.77

 EconPapers

-0.76

 betweenstory

-0.76

InjectAttribute

-0.73

MLLoader

-0.73

 propOrder

-0.71

POSITIVE LOGITS

0.40

↵

0.38

hot

0.36

 réunion

0.32

 réunions

0.31

 semelhante

0.30

 indépendante

0.30

 feest

0.30

 vorge

0.30

Activations Density 0.000%

garbled or non-alphanumeric characters

meeting/gathering specific groups

No Comments

No Known Activations