INDEX

Explanations

descriptions related to physical attributes or features

oai_token-act-pair · gpt-3.5-turbo

phrases related to physical attributes or features

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

apsed

-0.76

 Agenda

-0.76

 Declaration

-0.69

 jailed

-0.69

 Waterloo

-0.69

 Demand

-0.69

 Weeks

-0.66

 Rent

-0.66

 Lifetime

-0.66

ãĥ¯ãĥ³

-0.65

POSITIVE LOGITS

 accent

1.26

 aest

1.25

 detract

1.20

 visually

1.20

 confuse

1.20

 resemble

1.07

 evoke

1.03

 minimize

1.03

 simulate

1.02

 signify

1.02

Activations Density 0.415%

descriptions related to physical attributes or features

phrases related to physical attributes or features

No Comments

No Known Activations