INDEX

Explanations

proper nouns related to specific locations and entities

oai_token-act-pair · gpt-3.5-turbo

names of countries, regions, and specific groups or entities

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

local

-0.75

staking

-0.70

oteric

-0.67

estinal

-0.65

olphins

-0.62

orbit

-0.62

 Fukushima

-0.61

igious

-0.61

 DRAGON

-0.60

Rocket

-0.60

POSITIVE LOGITS

 Syndrome

0.85

 Forever

0.73

topia

0.68

 syndrome

0.67

 Facts

0.67

tics

0.67

 Encyclopedia

0.66

 broch

0.66

hood

0.64

 whis

0.64

Activations Density 0.475%

proper nouns related to specific locations and entities

names of countries, regions, and specific groups or entities

No Known Activations