INDEX

Explanations

words related to physical objects and body parts

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Ãľ

-0.64

udeb

-0.62

employment

-0.61

EMP

-0.60

 Forth

-0.60

 Spac

-0.60

~~~~~~~~~~~~~~~~

-0.60

 Freak

-0.59

 SPACE

-0.58

Gos

-0.58

POSITIVE LOGITS

bell

1.31

ular

0.96

iard

0.89

south

0.87

ows

0.86

sburgh

0.86

otine

0.85

ertodd

0.85

shaw

0.84

 bells

0.82

Activations Density 0.005%

words related to physical objects and body parts

No Comments

No Known Activations