INDEX

Explanations

terms related to computer programming and IT infrastructure

oai_token-act-pair · gpt-3.5-turbo

negative terms related to system functionality and errors in technical contexts

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

Hoo

-0.78

Hus

-0.74

 Kaiser

-0.74

 Volks

-0.74

Ö¼

-0.72

 fraternity

-0.71

 Metall

-0.71

 Polit

-0.70

 Phelps

-0.69

GUN

-0.69

POSITIVE LOGITS

compatible

1.37

mode

1.33

enabled

1.29

optim

1.28

level

1.24

interface

1.24

stack

1.23

readable

1.23

usable

1.23

layer

1.22

Activations Density 0.085%

terms related to computer programming and IT infrastructure

negative terms related to system functionality and errors in technical contexts

No Comments

No Known Activations