INDEX

Explanations

positive emotional descriptions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "warm" and its associated meanings, emphasizing a sense of comfort and positivity

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

argon

-0.80

 IMAGES

-0.78

issors

-0.72

DEV

-0.71

gur

-0.70

aurus

-0.70

ONE

-0.69

angler

-0.69

IBLE

-0.68

doms

-0.68

POSITIVE LOGITS

achine

1.27

 warm

1.05

est

1.02

 fuzz

0.98

 warmth

0.92

warm

0.87

ening

0.87

 warmer

0.86

 fuzzy

0.86

 welcome

0.86

Activations Density 0.013%

positive emotional descriptions

instances of the word "warm" and its associated meanings, emphasizing a sense of comfort and positivity

No Comments

No Known Activations