INDEX

Explanations

past tense verbs

oai_token-act-pair · gpt-3.5-turbo

statements related to accountability and consequences

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

ELS

-0.57

thodox

-0.55

Il

-0.52

los

-0.52

Bridge

-0.49

airs

-0.49

Eastern

-0.48

updated

-0.48

Balt

-0.48

Ń

-0.48

POSITIVE LOGITS

 yourself

1.30

 yourselves

1.19

 Yourself

0.89

 your

0.76

 YOUR

0.70

poke

0.67

your

0.63

Your

0.61

 panties

0.60

 smack

0.59

Activations Density 0.872%

past tense verbs

statements related to accountability and consequences

No Comments

No Known Activations