INDEX

Explanations

instances where ellipses (...) are used

oai_token-act-pair · gpt-3.5-turbo

expressions of uncertainty and speculation

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

 Trail

-0.68

 Lama

-0.68

 dome

-0.64

stice

-0.64

soever

-0.64

 grate

-0.63

wagen

-0.63

 happiest

-0.63

 Pistons

-0.63

ebus

-0.63

POSITIVE LOGITS

normal

0.75

demand

0.70

certain

0.69

undreds

0.68

coming

0.67

their

0.66

them

0.66

polit

0.66

those

0.66

traditional

0.65

Activations Density 0.382%

instances where ellipses (...) are used

expressions of uncertainty and speculation

No Comments

No Known Activations