INDEX

Explanations

phrases that suggest possibility or uncertainty

oai_token-act-pair · gpt-3.5-turbo

tentative or conditional language, particularly the word "might" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

oric

-0.83

cies

-0.70

 Dragonbound

-0.70

rett

-0.68

uctor

-0.68

*/(

-0.66

 Lauder

-0.64

cedented

-0.63

ials

-0.63

pires

-0.62

POSITIVE LOGITS

 someday

1.05

haps

0.91

 conce

0.91

 feas

0.86

aea

0.85

 plaus

0.83

ily

0.82

be

0.81

iest

0.80

 mistakenly

0.75

Activations Density 0.036%

phrases that suggest possibility or uncertainty

tentative or conditional language, particularly the word "might" and its variations

No Comments

No Known Activations