INDEX

Explanations

That specific author that appears in text

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 SetTextColor

-0.74

 Ensure

-0.63

Gap

-0.62

 Skydragon

-0.62

 Tactics

-0.60

 Highlights

-0.60

 manifests

-0.60

 Stability

-0.60

å£«

-0.60

pedia

-0.59

POSITIVE LOGITS

agine

1.27

 glad

1.13

 gonna

1.07

 thankful

1.04

 grateful

1.03

 proud

0.99

 guessing

0.98

 fortunate

0.92

 ashamed

0.90

mediately

0.89

Activations Density 0.057%

That specific author that appears in text

No Comments

No Known Activations