INDEX

Explanations

paragraphs expressing accountability or responsibility

oai_token-act-pair · gpt-3.5-turbo

articles and their frequency in the context of various subjects

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Edit

-0.87

errors

-0.79

encies

-0.78

fn

-0.77

AIDS

-0.76

 denies

-0.76

issues

-0.72

boats

-0.72

EMA

-0.72

edit

-0.71

POSITIVE LOGITS

 culmination

1.05

 mere

1.01

 whopping

1.01

 huge

0.96

 reminder

0.96

 continuation

0.96

 mixture

0.95

 thinly

0.95

 relatively

0.92

 conglomer

0.92

Activations Density 0.276%

paragraphs expressing accountability or responsibility

articles and their frequency in the context of various subjects

No Comments

No Known Activations