INDEX

Explanations

phrases related to receiving emails from a specific source

oai_token-act-pair · gpt-3.5-turbo

prepositions and phrases indicating relationships or connections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

po

-0.77

bol

-0.73

tip

-0.69

birth

-0.68

="#

-0.67

witz

-0.67

bara

-0.66

etsk

-0.66

ItemTracker

-0.65

camp

-0.64

POSITIVE LOGITS

 alike

1.66

 respectively

1.61

depending

0.96

 whichever

0.85

 respective

0.79

 interchange

0.74

each

0.71

 thereof

0.69

 both

0.68

 interchangeable

0.67

Activations Density 0.857%

phrases related to receiving emails from a specific source

prepositions and phrases indicating relationships or connections

No Comments

No Known Activations