INDEX

Explanations

the word "pie"

oai_token-act-pair · gpt-3.5-turbo

references to "pie" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

spring

-0.83

 Students

-0.74

nen

-0.71

iveness

-0.71

nesses

-0.68

runners

-0.66

 EDITION

-0.65

wcsstore

-0.64

ities

-0.63

sound

-0.63

POSITIVE LOGITS

pie

1.33

 pies

1.23

pie

1.08

Pie

0.94

Pie

0.94

 crust

0.90

 desserts

0.85

cing

0.82

oshenko

0.81

artisan

0.81

Activations Density 0.007%

the word "pie"

references to "pie" in various contexts

No Comments

No Known Activations