INDEX

Explanations

quotes within text

oai_token-act-pair · gpt-3.5-turbo

quotation marks and dialogue

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

 bunk

-0.76

 favor

-0.75

 honor

-0.71

 shack

-0.66

 veget

-0.65

 slam

-0.64

 footing

-0.64

 monog

-0.64

fab

-0.63

 annually

-0.63

POSITIVE LOGITS

Therefore

1.09

Whereas

1.05

Fortunately

1.02

Moreover

1.01

There

0.98

Certainly

0.97

However

0.96

We

0.95

Nevertheless

0.95

Whoever

0.94

Activations Density 0.068%

quotes within text

quotation marks and dialogue

No Comments

No Known Activations