INDEX

Explanations

legal or official terms and procedures

oai_token-act-pair · gpt-3.5-turbo

phrases related to intent or purpose in actions

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

 Flavoring

-0.63

è¦ļéĨĴ

-0.62

merce

-0.61

terness

-0.61

 Enhancement

-0.61

)]

-0.59

 Interstellar

-0.59

 Transparency

-0.58

"],"

-0.57

]).

-0.56

POSITIVE LOGITS

 ensued

0.66

 billed

0.63

 startled

0.63

 afterward

0.62

 locals

0.60

 subsequently

0.60

 greeted

0.59

 awoke

0.58

 followed

0.58

 horrified

0.58

Activations Density 1.405%

legal or official terms and procedures

phrases related to intent or purpose in actions

No Comments

No Known Activations