INDEX

Explanations

numerical values and their associated context in text

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Autorizaciones

-0.52

Jereo

-0.45

этому

-0.40

 OMITBAD

-0.39

AccessorTable

-0.38

 kaarangay

-0.38

WriteBarrier

-0.36

BindingResult

-0.35

inheritdoc

-0.34

 Bioaccumulative

-0.34

POSITIVE LOGITS

transQ

0.54

 BoxFit

0.52

 CreateTagHelper

0.51

*/),

0.51

 cherchés

0.50

 Bucure

0.49

 للاسماء

0.49

 ModelExpression

0.46

BASELINE

0.45

ükemmel

0.44

Activations Density 0.094%

numerical values and their associated context in text

No Comments

No Known Activations