INDEX

Explanations

references to various forms of work or employment

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

"work" or "Work"

np_acts-logits-general · gemini-2.0-flash

work continuations

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

istoitu

-0.65

 pingente

-0.63

 cadeia

-0.60

 capuz

-0.56

 namorados

-0.52

 Siegel

-0.51

 mídia

-0.50

orghini

-0.50

 Galbraith

-0.50

<bos>

-0.49

POSITIVE LOGITS

Work

1.17

 Work

1.09

 work

1.05

work

1.04

WORK

0.96

 WORK

0.94

Working

0.94

working

0.93

 working

0.90

 Working

0.84

Activations Density 0.128%

references to various forms of work or employment

"work" or "Work"

work continuations

No Comments

No Known Activations