INDEX

Explanations

references to quantities, such as numbers and measurements

oai_token-act-pair · gpt-3.5-turbo

phrases related to potential risks and medical 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

Nap

-0.62

podcast

-0.59

Baltimore

-0.58

undrum

-0.58

 Emblem

-0.58

 Patreon

-0.57

 Cannes

-0.57

NFL

-0.56

 Quinn

-0.55

 keynote

-0.55

POSITIVE LOGITS

)).

0.79

 undet

0.78

 carbohyd

0.76

)."

0.74

]."

0.74

 detectable

0.72

).[

0.70

'."

0.69

 attRot

0.68

 versa

0.68

Activations Density 2.059%

references to quantities, such as numbers and measurements

phrases related to potential risks and medical issues

No Comments

No Known Activations