INDEX

Explanations

names, abbreviations, and trademarked product names

oai_token-act-pair · gemini-2.0-flash

names Method used: 1 — because max-activating tokens are proper names — names

np_max-act-logits · gpt-5-mini Triggered by @chenshw0109

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_4/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 ujednoznacz

-0.72

reportWebVitals

-0.66

Filmografie

-0.65

igshid

-0.65

-0.59

例文帳に追加

-0.59

Viitteet

-0.57

($.

-0.56

 CURIAM

-0.56

 canst

-0.56

POSITIVE LOGITS

<bos>

2.16

uxxxx

0.59

0.52

0.51

openConnection

0.49

’

0.48

0.47

0.46

 '\\;'

0.46

Activations Density 15.733%

names, abbreviations, and trademarked product names

names Method used: 1 — because max-activating tokens are proper names — names

No Comments

No Known Activations