INDEX

Explanations

words related to strong emotions or sentiments, such as being saddened, disappointed, or appalled

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

gravity

-0.62

zig

-0.61

Ranked

-0.60

 nutshell

-0.58

illary

-0.57

aceutical

-0.57

essage

-0.57

ua

-0.57

entials

-0.56

zeb

-0.55

POSITIVE LOGITS

by

1.27

 aback

1.05

by

1.01

By

0.93

BY

0.92

By

0.84

dy

0.80

 saddened

0.73

 Dragonbound

0.72

 anew

0.71

Activations Density 0.109%

words related to strong emotions or sentiments, such as being saddened, disappointed, or appalled

No Comments

No Known Activations