INDEX

Explanations

proper nouns "Sol" appearing in the text

oai_token-act-pair · gpt-3.5-turbo

references to the word "Sol" and its various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

shapeshifter

-0.80

è¦ļéĨĴ

-0.72

SHIP

-0.70

 Ancients

-0.68

IRD

-0.65

 conscious

-0.65

alm

-0.63

 healthy

-0.60

ulhu

-0.60

reader

-0.60

POSITIVE LOGITS

itary

1.15

ving

1.03

stice

1.02

ved

1.01

omon

0.98

vable

0.96

Sol

0.92

itaire

0.92

oshenko

0.89

ven

0.87

Activations Density 0.008%

proper nouns "Sol" appearing in the text

references to the word "Sol" and its various contexts

No Comments

No Known Activations