INDEX

Explanations

sections and references to additional information or comments in a document

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

like "notes", "comments", or "remarks"

np_acts-logits-general · gemini-2.0-flash

notes, comments, and references

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

Discografia

-0.56

andExpect

-0.50

存于互联网档案馆

-0.50

 GenerationType

-0.50

DockStyle

-0.49

Gön

-0.48

蚪

-0.47

unhofer

-0.45

----</

-0.45

 Мексичка

-0.45

POSITIVE LOGITS

 notes

0.61

 Notes

0.61

Notes

0.60

 Bemerkungen

0.53

notes

0.53

 Notas

0.52

 note

0.51

 NOTES

0.50

 referenties

0.47

 Note

0.47

Activations Density 0.378%

sections and references to additional information or comments in a document

like "notes", "comments", or "remarks"

notes, comments, and references

No Comments

No Known Activations