INDEX

Explanations

questions or statements directed towards the reader

oai_token-act-pair · gpt-3.5-turbo

questions directed at the reader that encourage engagement

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

currently

-0.71

Maker

-0.71

anon

-0.67

rette

-0.66

arget

-0.65

objects

-0.65

renheit

-0.65

assemb

-0.64

ilus

-0.64

Member

-0.64

POSITIVE LOGITS

 stumble

0.77

 lapse

0.73

arcity

0.72

 looph

0.72

 omission

0.71

 earlier

0.70

 foresee

0.70

ACP

0.69

 last

0.69

 bailed

0.69

Activations Density 0.244%

questions or statements directed towards the reader

questions directed at the reader that encourage engagement

No Comments

No Known Activations