INDEX

Explanations

terms related to technical discussions, such as "reference," "opinions," and "test."

oai_token-act-pair · gpt-3.5-turbo

references to personal work or projects

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

 ÂŃ

-1.00

etooth

-0.79

Mobil

-0.77

ï¿½ï¿½

-0.71

 leased

-0.71

ÂŃ

-0.71

apeake

-0.69

 homeowners

-0.67

ordable

-0.66

 JPMorgan

-0.64

POSITIVE LOGITS

 spoilers

1.38

 spoiler

1.30

 anime

1.23

 Anime

1.20

 manga

1.14

Characters

1.13

 canon

1.10

 sequels

1.08

 Characters

1.06

 playthrough

1.04

Activations Density 0.867%

terms related to technical discussions, such as "reference," "opinions," and "test."

references to personal work or projects

No Comments

No Known Activations