INDEX

Explanations

terms that indicate variation or different levels of something

oai_token-act-pair · gpt-3.5-turbo

concepts of variation in characteristics or qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

stra

-0.84

vest

-0.75

hip

-0.73

talk

-0.73

 Printing

-0.73

hra

-0.72

paper

-0.70

WARD

-0.69

abad

-0.69

hers

-0.69

POSITIVE LOGITS

iating

0.99

 shades

0.94

 degrees

0.92

 amounts

0.92

iated

0.91

iates

0.87

 variations

0.86

 combinations

0.86

 lengths

0.86

 conting

0.86

Activations Density 0.017%

terms that indicate variation or different levels of something

concepts of variation in characteristics or qualities

No Comments

No Known Activations