INDEX

Explanations

phrases or descriptions focusing on small proportions or minorities within a larger group

oai_token-act-pair · gpt-3.5-turbo

references to small or tiny quantities or groups

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

ources

-0.78

ourses

-0.76

YR

-0.72

ses

-0.69

sem

-0.68

chwitz

-0.67

emis

-0.67

sbm

-0.66

aucus

-0.65

yrs

-0.64

POSITIVE LOGITS

pox

1.31

 handful

1.06

 amount

1.05

 fraction

1.02

 percentage

1.01

 subset

0.89

 portion

0.87

 snippet

0.87

 footprint

0.87

 sample

0.86

Activations Density 0.052%

phrases or descriptions focusing on small proportions or minorities within a larger group

references to small or tiny quantities or groups

No Comments

No Known Activations