INDEX

Explanations

terms related to legal and governmental contexts

oai_token-act-pair · gemini-2.0-flash

adjectives ending in -ing

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Doing

-0.60

 betweenstory

-0.56

 doing

-0.54

Working

-0.54

 Working

-0.54

 ModelExpression

-0.52

Doing

-0.50

 eingeladen

-0.48

 working

-0.46

 Taking

-0.46

POSITIVE LOGITS

 challenging

1.23

 demanding

1.15

 threatening

1.06

 inspiring

1.02

 rewarding

1.02

 distracting

1.02

 Challenging

1.02

 promising

0.98

 discouraging

0.98

 compelling

0.95

Activations Density 15.452%

terms related to legal and governmental contexts

adjectives ending in -ing

No Comments

No Known Activations