INDEX

Explanations

questions, particularly negative or challenging questions often with "don't"

oai_token-act-pair · gemini-2.0-flash

t

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_7/width_16k/average_l0_70

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 MainAxisSize

-0.87

ftagPool

-0.84

enumi

-0.80

IntoConstraints

-0.79

AndEndTag

-0.79

 للاسماء

-0.77

parsedMessage

-0.75

setVerticalGroup

-0.73

AnchorStyles

-0.72

 ModelExpression

-0.72

POSITIVE LOGITS

removeItem

0.49

allclose

0.47

<eos>

0.47

 formerly

0.44

 diverse

0.44

 restitu

0.43

riko

0.43

を持ち

0.43

Lato

0.42

 Réunion

0.42

Activations Density 4.539%

questions, particularly negative or challenging questions often with "don't"

t

No Comments

No Known Activations