INDEX

Explanations

phrases that indicate action completion

oai_token-act-pair · gpt-3.5-turbo

the presence of sentences and punctuation

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

 enroll

-0.85

 credential

-0.83

 emanc

-0.82

 authorized

-0.80

 privileged

-0.80

 enrolled

-0.79

 enrol

-0.79

eteen

-0.78

 distribut

-0.78

 invitation

-0.77

POSITIVE LOGITS

 Especially

1.37

 Whilst

1.33

 Hopefully

1.32

 Basically

1.32

;)

1.31

 Anyway

1.31

:)

1.29

 Secondly

1.28

 Unfortunately

1.27

 Otherwise

1.27

Activations Density 0.907%

phrases that indicate action completion

the presence of sentences and punctuation

No Comments

No Known Activations