INDEX

Explanations

phrases related to iconic or unique concepts or names

oai_token-act-pair · gpt-3.5-turbo

titles of media or content related to notable works or artists

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

haps

-0.85

lessly

-0.84

orously

-0.80

aciously

-0.80

cific

-0.78

 disg

-0.77

ulates

-0.77

aez

-0.76

istration

-0.76

matically

-0.76

POSITIVE LOGITS

 Mind

1.05

 Places

1.00

 Dreams

1.00

 Things

1.00

 Ones

1.00

 Hands

0.99

 Tide

0.99

 Feet

0.97

 Limits

0.95

 Worlds

0.94

Activations Density 0.155%

phrases related to iconic or unique concepts or names

titles of media or content related to notable works or artists

No Comments

No Known Activations