INDEX

Explanations

details related to medical emergencies or incidents happening in a hospital emergency room

oai_token-act-pair · gpt-3.5-turbo

references to past events or experiences

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

).[

-0.67

anwhile

-0.66

"/>

-0.66

)."

-0.65

 tuber

-0.62

 endif

-0.62

 ster

-0.60

.'"

-0.60

'."

-0.58

 obstruct

-0.57

POSITIVE LOGITS

bernatorial

0.83

atform

0.73

podcast

0.70

odcast

0.64

 Podcast

0.62

widget

0.59

 Awards

0.58

ãĤ¢ãĥ«

0.58

è£ħ

0.57

NAS

0.57

Activations Density 1.683%

details related to medical emergencies or incidents happening in a hospital emergency room

references to past events or experiences

No Comments

No Known Activations