INDEX

Explanations

mentions of events happening to people in the past.

oai_token-act-pair · gemini-2.0-flash

past events

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_19/width_16k/average_l0_12

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.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

was

-1.20

had

-0.98

 appeared

-0.94

 seemed

-0.94

was

-0.90

 were

-0.90

 consisted

-0.85

 wasn

-0.84

 differed

-0.81

Was

-0.79

POSITIVE LOGITS

DeleteBehavior

0.77

]--;

0.68

DockStyle

0.59

abond

0.57

Билгалдахарш

0.56

پیوند

0.56

Cooperative

0.56

Française

0.55

 jScrollPane

0.54

mbor

0.53

Activations Density 7.297%

mentions of events happening to people in the past.

past events

No Comments

No Known Activations