INDEX

Explanations

phrases indicating a significant or historical event happening for the first time

oai_token-act-pair · gpt-3.5-turbo

phrases indicating significant historical events or first occurrences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

asty

-0.83

 tomat

-0.69

phen

-0.67

CHAT

-0.65

lua

-0.64

bern

-0.63

peed

-0.62

krit

-0.62

pes

-0.61

urous

-0.60

POSITIVE LOGITS

ever

1.00

 ever

0.98

 EVER

0.96

 imaginable

0.89

Ever

0.74

 anniversary

0.73

 since

0.70

 Scotia

0.67

 Ever

0.64

 practicable

0.62

Activations Density 0.035%

phrases indicating a significant or historical event happening for the first time

phrases indicating significant historical events or first occurrences

No Comments

No Known Activations