INDEX

Explanations

names of people and associated actions or situations involving those people

oai_token-act-pair · gpt-3.5-turbo

references to individuals or entities involved in legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

rossover

-0.79

 Rankings

-0.71

apters

-0.70

 Miko

-0.69

STEM

-0.67

urban

-0.66

 apocalyptic

-0.65

romeda

-0.64

pop

-0.63

 Dise

-0.63

POSITIVE LOGITS

 complied

1.15

 voluntarily

1.13

 violated

1.12

 complying

1.03

 lawfully

1.00

 waived

0.98

 unlawfully

0.98

 denied

0.97

 disclaim

0.97

 misled

0.96

Activations Density 0.919%

names of people and associated actions or situations involving those people

references to individuals or entities involved in legal contexts

No Comments

No Known Activations