INDEX

Explanations

conditional statements about people in hypothetical situations

oai_token-act-pair · gemini-2.0-flash

future tense

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 können

-0.82

can

-0.81

 dapat

-0.78

 kann

-0.75

 pouvons

-0.73

Can

-0.71

Puedes

-0.71

 Pueden

-0.68

can

-0.68

可

-0.65

POSITIVE LOGITS

 will

1.46

will

1.30

Will

1.13

 Will

1.09

 akan

0.95

WILL

0.94

 WILL

0.92

 sẽ

0.91

 would

0.82

 θα

0.76

Activations Density 1.662%

conditional statements about people in hypothetical situations

future tense

No Comments

No Known Activations