INDEX

Explanations

words related to being mentioned or recorded in a list or catalog

oai_token-act-pair · gpt-3.5-turbo

instances of the word "listed" in 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

qua

-0.83

issance

-0.75

pling

-0.67

ggles

-0.67

omen

-0.66

isoft

-0.66

ipal

-0.65

assi

-0.64

Pear

-0.64

Measure

-0.64

POSITIVE LOGITS

 listing

1.01

 listings

0.88

erv

0.85

onyms

0.84

 listed

0.81

witz

0.80

 lists

0.75

icter

0.72

 widget

0.71

daq

0.70

Activations Density 0.016%

words related to being mentioned or recorded in a list or catalog

instances of the word "listed" in various contexts

No Comments

No Known Activations