INDEX

Explanations

instances where actions may result in consequences, such as being removed, arrested, judged, or eliminated

oai_token-act-pair · gpt-3.5-turbo

instances of the word "and."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

hov

-0.78

rain

-0.72

anza

-0.72

hof

-0.69

laus

-0.68

ept

-0.67

Morning

-0.67

itans

-0.66

 Doodle

-0.66

âĻ¥

-0.66

POSITIVE LOGITS

 hence

1.04

 therefore

1.00

 consequently

0.98

 experimented

0.96

 repaired

0.91

 subsequently

0.90

 thus

0.89

 redistributed

0.88

 debated

0.86

 then

0.85

Activations Density 0.270%

instances where actions may result in consequences, such as being removed, arrested, judged, or eliminated

instances of the word "and."

No Comments

No Known Activations