INDEX

Explanations

specific entities or products mentioned in the news

oai_token-act-pair · gpt-3.5-turbo

keywords related to various types of products and entities

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

Negative Logits

 Helpful

-0.90

 largeDownload

-0.75

 oneself

-0.72

odder

-0.69

aples

-0.69

oln

-0.68

renches

-0.68

Beg

-0.67

cale

-0.67

eger

-0.66

POSITIVE LOGITS

 counterparts

0.91

mates

0.87

 registrations

0.84

 woes

0.83

 partners

0.79

mate

0.78

 mates

0.77

 replacements

0.77

 namesake

0.76

 prowess

0.76

Activations Density 0.504%

specific entities or products mentioned in the news

keywords related to various types of products and entities

No Known Activations