INDEX

Explanations

terms related to information security and privacy concerns

oai_token-act-pair · gpt-3.5-turbo

phrases related to social awareness and activism issues

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

)|

-0.48

externalActionCode

-0.47

))))

-0.47

 Technique

-0.45

REDACTED

-0.45

 Ultron

-0.43

 annex

-0.42

Ö¼

-0.42

Ves

-0.42

 Dragonbound

-0.41

POSITIVE LOGITS

 themselves

0.70

 their

0.68

their

0.60

 THEIR

0.56

 nowadays

0.49

agonists

0.49

 histories

0.48

 varying

0.48

chwitz

0.47

 variations

0.47

Activations Density 2.142%

terms related to information security and privacy concerns

phrases related to social awareness and activism issues

No Comments

No Known Activations