INDEX

Explanations

terms related to expectations or predictions

oai_token-act-pair · gpt-3.5-turbo

terms related to expectation or prediction

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

undy

-0.79

bard

-0.78

yards

-0.74

hide

-0.73

ggies

-0.73

ena

-0.69

bag

-0.68

eled

-0.68

eno

-0.67

vard

-0.66

POSITIVE LOGITS

 anticip

0.91

 anticipation

0.89

 foresee

0.84

 anticipating

0.80

 anticipate

0.79

ROR

0.79

 anticipated

0.74

eers

0.72

aneously

0.69

noon

0.66

Activations Density 0.018%

terms related to expectations or predictions

terms related to expectation or prediction

No Comments

No Known Activations