INDEX

Explanations

datelike strings, such as numbers, times, numbered lists, and fragmented dates

oai_token-act-pair · gemini-2.0-flash

legal document

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

 chèvre

-0.66

 agré

-0.65

 quæ

-0.64

 occidentalis

-0.63

tagHelperRunner

-0.63

gettimeofday

-0.62

 înainte

-0.62

 <<<<<<<<<<<<<<

-0.60

 незавершена

-0.60

BlockingQueue

-0.59

POSITIVE LOGITS

 AssemblyTitle

0.60

 }}"></

0.44

CONCLUSION

0.42



0.41

 Mars

0.40

SENIOR

0.40

❑

0.39

RESEARCH

0.38

 intptr

0.38



0.37

Activations Density 11.272%

datelike strings, such as numbers, times, numbered lists, and fragmented dates

legal document

No Comments

No Known Activations