INDEX

Explanations

words related to discussing and deliberating on various topics or issues

oai_token-act-pair · gpt-3.5-turbo

references to discussions and issues being addressed

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

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

VERT

-0.56

 Gamble

-0.54

Ham

-0.53

 Carb

-0.53

SPONSORED

-0.52

ilts

-0.52

 ignition

-0.52

enta

-0.51

é¾įå

-0.51

 Morton

-0.51

POSITIVE LOGITS

 differently

1.34

 anew

1.22

 aloud

1.20

 separately

1.17

 respectfully

1.11

 accurately

1.09

 responsibly

1.07

 prominently

1.06

 correctly

1.05

 accordingly

1.05

Activations Density 0.647%

words related to discussing and deliberating on various topics or issues

references to discussions and issues being addressed

No Comments

No Known Activations