INDEX

Explanations

technical or scientific specific concepts, including manufacturing, electronic components, network systems. Also finds legal terminology ("mistrial", etc), calculated results like exam marks, and reproductive biology.

oai_token-act-pair · gemini-2.0-flash

Jargon

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

en

-0.48

gnus

-0.45

ven

-0.45

idor

-0.41

icos

-0.40

id

-0.40

Spawner

-0.39

lagt

-0.39

Kom

-0.39

POSITIVE LOGITS

 nakalista

0.83

 fubject

0.81

 Majefty

0.79

 Theſe

0.79

 purpoſe

0.78

ThroughAttribute

0.77

extAlignment

0.76

 kaarangay

0.76

ſelf

0.75

NUMX

0.72

Activations Density 4.097%

technical or scientific specific concepts, including manufacturing, electronic components, network systems. Also finds legal terminology ("mistrial", etc), calculated results like exam marks, and reproductive biology.

Jargon

No Comments

No Known Activations