INDEX

Explanations

phrases related to air travel and aviation

oai_token-act-pair · gpt-3.5-turbo

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

 recollection

-0.71

 Braz

-0.70

 initials

-0.70

 Hilbert

-0.68

 Beir

-0.68

ABE

-0.67

Phi

-0.66

bos

-0.66

 Triangle

-0.65

POSITIVE LOGITS

condition

1.29

borne

1.24

based

1.23

weather

1.13

tight

1.10

raid

1.10

surface

1.07

saving

1.06

floor

1.05

tested

1.05

Activations Density 0.056%

phrases related to air travel and aviation

No Comments

No Known Activations