INDEX

Explanations

specific mentions of dates and events

oai_token-act-pair · gpt-3.5-turbo

important dates and historical events

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

".[

-0.98

''.

-0.93

 recognise

-0.86

catentry

-0.82

 Whilst

-0.79

Others

-0.78

".

-0.77

)).

-0.77

 enqu

-0.77

 realise

-0.76

POSITIVE LOGITS

 Slate

0.75

 Kimmel

0.71

 Lafayette

0.67

 Whitman

0.66

 FedEx

0.65

 Dunham

0.64

 debuted

0.64

 Indianapolis

0.64

 Georgetown

0.63

 Philly

0.63

Activations Density 1.839%

specific mentions of dates and events

important dates and historical events

No Comments

No Known Activations