INDEX

Explanations

seems to be triggered by a mix of seemingly unrelated things: code snippets, locations, medical terms, and some common words/punctuation

oai_token-act-pair · gemini-2.0-flash

medical data

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_12/width_16k/average_l0_6

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

ViewImports

-1.02

ArgsConstructor

-0.96

 disambiguazione

-0.91

AndEndTag

-0.88

DockStyle

-0.88

EndGlobalSection

-0.88

 defaultstate

-0.87

HostException

-0.85

 незавершена

-0.84

 noqa

-0.82

POSITIVE LOGITS

 pele

0.44

 back

0.42

per

0.41

mo

0.40

me

0.39

“

0.39

 given

0.38

menti

0.37

 direct

0.37

trans

0.37

Activations Density 1.723%

seems to be triggered by a mix of seemingly unrelated things: code snippets, locations, medical terms, and some common words/punctuation

medical data

No Comments

No Known Activations

seems to be triggered by a mix of seemingly unrelated things: code snippets, locations, medical terms, and some common words/punctuation

medical data

No Comments

No Known Activations