INDEX

Explanations

phrases related to societal or political issues, particularly with a critical perspective

oai_token-act-pair · gpt-3.5-turbo

references to a shared community or collective experience

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

bender

-0.76

ingle

-0.74

ouse

-0.73

puff

-0.72

wrap

-0.72

yang

-0.72

quart

-0.70

edit

-0.70

 Levine

-0.68

edi

-0.67

POSITIVE LOGITS

selves

1.25

 ancestors

1.21

own

1.13

 nation

1.13

 shores

1.08

 society

1.06

 beloved

1.04

 selves

1.04

 hearts

1.03

 democracy

1.00

Activations Density 0.123%

phrases related to societal or political issues, particularly with a critical perspective

references to a shared community or collective experience

No Comments

No Known Activations