INDEX

Explanations

words related to computer programming languages and terms

oai_token-act-pair · gpt-3.5-turbo

occurrences of prefixes and suffixes related to words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

anish

-0.67

 torches

-0.63

 Bounty

-0.61

©¶æ

-0.61

agos

-0.60

Bey

-0.59

BuyableInstoreAndOnline

-0.58

Ïģ

-0.58

 Allaah

-0.58

hospital

-0.58

POSITIVE LOGITS

swick

0.97

enhagen

0.79

ãĥ¼ãĤ¯

0.78

ittees

0.70

ãĥŃ

0.70

geist

0.67

ãĥĥãĥī

0.67

ãĥŁ

0.67

enment

0.65

ensical

0.64

Activations Density 0.100%

words related to computer programming languages and terms

occurrences of prefixes and suffixes related to words

No Comments

No Known Activations