INDEX

Explanations

descriptive action during a car chase

oai_token-act-pair · gemini-2.0-flash

essentially garbage/meaningless inputs

np_token-act-pair-logits · gemini-2.0-flash

accidents

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_16/width_16k/average_l0_10

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

 cintur

-0.56

munic

-0.55

präsident

-0.52

schlag

-0.51

 greateſt

-0.50

Ifc

-0.50

esercito

-0.50

 Agricultura

-0.49

akat

-0.49

 engraçadas

-0.49

POSITIVE LOGITS

 MenuView

0.57

 חיצוניים

0.52

 AssemblyProduct

0.50

zeera

0.49

 Sinon

0.47

lipop

0.47

FormTagHelper

0.47

stě

0.47

exitRule

0.46

atshirt

0.46

Activations Density 1.947%

descriptive action during a car chase

essentially garbage/meaningless inputs

accidents

No Comments

No Known Activations