INDEX

Explanations

phrases related to media and news stories

oai_token-act-pair · gpt-3.5-turbo

commas in the text

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

¬¼

-0.74

ety

-0.63

vered

-0.62

ocl

-0.62

enary

-0.60

UF

-0.58

iple

-0.58

irie

-0.57

isi

-0.57

uber

-0.57

POSITIVE LOGITS

 albeit

1.15

 although

1.07

 whereas

1.03

 namely

0.97

but

0.96

 which

0.96

 though

0.93

 including

0.93

 regardless

0.86

 whereby

0.85

Activations Density 0.721%

phrases related to media and news stories

commas in the text

No Comments

No Known Activations