INDEX

Explanations

phrases related to optimization and maximizing efficiency or performance

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_32k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.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

 betweenstory

-0.59

lappe

-0.54

setOrder

-0.53

indak

-0.50

 resourceCulture

-0.50

Postup

-0.50

IndentedString

-0.49

 виправивши

-0.49

 cough

-0.49

тером

-0.48

POSITIVE LOGITS

 maximize

1.13

 maximizing

1.07

 maximizes

0.99

maximize

0.98

 maximise

0.97

 maximized

0.96

 optimum

0.94

 Maximize

0.93

 optimal

0.90

 optimize

0.88

Activations Density 0.273%

phrases related to optimization and maximizing efficiency or performance

No Comments

No Known Activations