INDEX

Explanations

terms related to approximation and approximation methods

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

RenderAtEndOf

-0.64

esModule

-0.60

@[+][

-0.57

 stazione

-0.52

SharedDtor

-0.48

Famille

-0.48

sendRedirect

-0.48

setTextColor

-0.48

getContent

-0.47

LEGGI

-0.47

POSITIVE LOGITS

 approximation

0.92

 Approximation

0.81

 approximate

0.81

ximation

0.80

 Approximate

0.80

Approximate

0.77

 aproxima

0.77

 approximates

0.75

 approximations

0.72

 APPRO

0.71

Activations Density 0.024%

terms related to approximation and approximation methods

No Comments

No Known Activations