INDEX

Explanations

characters that appear to be separators or special characters within a text

oai_token-act-pair · gpt-3.5-turbo

instances of uncertainty or skepticism

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 eleph

-0.89

 undermin

-0.75

 Compass

-0.68

 constit

-0.66

 adversaries

-0.66

 proport

-0.64

 warships

-0.63

 tremend

-0.62

 representation

-0.60

 advisors

-0.60

POSITIVE LOGITS

posted

1.08

Posted

1.02

Anonymous

1.01

Anyway

0.96

↵

0.93

lol

0.91

Í

0.90

Honestly

0.89

EDIT

0.82

Seriously

0.82

Activations Density 0.278%

characters that appear to be separators or special characters within a text

instances of uncertainty or skepticism

No Comments

No Known Activations