INDEX

Explanations

quotations inside double quotation marks

oai_token-act-pair · gpt-3.5-turbo

quoted speech

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 prelim

-0.73

 livestream

-0.68

 gossip

-0.66

 filib

-0.66

clo

-0.65

 mids

-0.65

 spar

-0.65

 footing

-0.63

est

-0.62

 grades

-0.62

POSITIVE LOGITS

We

1.06

Certainly

0.97

Our

0.96

It

0.96

Interest

0.94

Very

0.93

There

0.93

icious

0.92

They

0.92

0.91

Activations Density 0.105%

quotations inside double quotation marks

quoted speech

No Comments

No Known Activations