INDEX

Explanations

quotations followed by statements of fact

oai_token-act-pair · gpt-3.5-turbo

dialogue or quotes attributed to speakers or sources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

mast

-0.76

ingu

-0.75

 contrace

-0.74

 enchanted

-0.66

ptin

-0.66

©¶æ

-0.65

 furnace

-0.64

Widget

-0.64

 Heavenly

-0.63

 helicop

-0.62

POSITIVE LOGITS

 reads

1.14

 writes

1.01

 wrote

0.99

 according

0.90

according

0.82

 concludes

0.76

said

0.74

 advises

0.73

 stated

0.73

 commented

0.73

Activations Density 0.144%

quotations followed by statements of fact

dialogue or quotes attributed to speakers or sources

No Comments

No Known Activations