INDEX

Explanations

letters and types of statistical or numerical values within logical propositions or equations

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

ckså

-0.53

RenderAtEndOf

-0.52

IsMutable

-0.51

ंदीखरीदारी

-0.50

OrBuilder

-0.49

izational

-0.49

PerformLayout

-0.48

хіво

-0.48

⅘

-0.47

 незавершена

-0.46

POSITIVE LOGITS

 <<<<<<<<<<<<<<

0.50

BeginContext

0.39

 maxim

0.39

Bal

0.39

0.36

maxim

0.36

térature

0.35

■

0.35

 Belmont

0.35

Meter

0.35

Activations Density 0.006%

letters and types of statistical or numerical values within logical propositions or equations

No Comments

No Known Activations