INDEX

Explanations

phrases related to personal accounts or stories

oai_token-act-pair · gpt-3.5-turbo

references to personal experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

inately

-0.72

nda

-0.70

efficiency

-0.68

leaf

-0.68

laws

-0.67

nod

-0.67

vous

-0.66

pillar

-0.65

corn

-0.65

gem

-0.64

POSITIVE LOGITS

 experiences

0.99

 experien

0.95

 firsthand

0.95

 Exper

0.87

 experience

0.86

 Experience

0.82

ually

0.82

iences

0.80

ional

0.75

Shape

0.72

Activations Density 0.036%

phrases related to personal accounts or stories

references to personal experiences

No Comments

No Known Activations