INDEX

Explanations

expressions related to personal viewpoints and opinions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating personal thoughts and opinions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Droid

-0.68

lur

-0.60

 Â®

-0.59

PDF

-0.59

 Contains

-0.58

 quir

-0.58

Blog

-0.58

 Released

-0.57

aily

-0.55

 Warranty

-0.54

POSITIVE LOGITS

entimes

1.18

 secondly

1.13

 yeah

1.09

 everybody

0.94

0.90

Secondly

0.90

 frankly

0.89

 somebody

0.89

 luckily

0.88

 Secondly

0.88

Activations Density 0.287%

expressions related to personal viewpoints and opinions

phrases indicating personal thoughts and opinions

No Comments

No Known Activations