INDEX

Explanations

sentences that highlight important points or suggestions in a discussion

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 judiciaire

-0.54

 mijne

-0.54

endphp

-0.52

MLLoader

-0.50

DockStyle

-0.49

 surla

-0.48

 ModelExpression

-0.48

SaveChangesAsync

-0.47

 snippetHide

-0.47

 numériques

-0.47

POSITIVE LOGITS

Personendaten

0.40

 valid

0.40

 good

0.39

NonNull

0.38

Haut

0.37

Good

0.37

good

0.36

 clue

0.35

angerie

0.35

 Prud

0.34

Activations Density 0.012%

sentences that highlight important points or suggestions in a discussion

No Comments

No Known Activations