INDEX

Explanations

phrases indicating the need for confirmation or validation

oai_token-act-pair · gpt-3.5-turbo

possessive pronouns and references to personal identity or membership

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Shutterstock

-0.80

backs

-0.74

soever

-0.69

 Springer

-0.69

 hoops

-0.68

back

-0.67

 paced

-0.67

bourg

-0.66

erton

-0.66

 Canaver

-0.65

POSITIVE LOGITS

 existence

1.44

 authenticity

1.37

 validity

1.33

 legitimacy

1.22

 innocence

1.17

worthiness

1.04

 presence

1.01

iability

0.96

 importance

0.96

 accuracy

0.96

Activations Density 0.247%

phrases indicating the need for confirmation or validation

possessive pronouns and references to personal identity or membership

No Comments

No Known Activations