INDEX

Explanations

questions or statements involving someone being asked about a specific topic

oai_token-act-pair · gpt-3.5-turbo

instances of the word "asked" indicating inquiries or requests for information

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

marine

-0.68

 Confeder

-0.63

Right

-0.62

Sham

-0.62

Est

-0.62

rats

-0.61

 Cock

-0.61

bows

-0.61

Liber

-0.60

ARM

-0.60

POSITIVE LOGITS

 rhet

1.03

 asked

0.91

govtrack

0.89

naires

0.84

 questions

0.83

ioned

0.82

 permission

0.79

 probing

0.77

 questioned

0.77

 repeatedly

0.74

Activations Density 0.018%

questions or statements involving someone being asked about a specific topic

instances of the word "asked" indicating inquiries or requests for information

No Comments

No Known Activations