INDEX

Explanations

famous individuals, specifically celebrities and athletes

oai_token-act-pair · gpt-3.5-turbo

terms related to celebrity or notable individuals 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 @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

mbuds

-0.74

noon

-0.73

ipedia

-0.66

aya

-0.65

 Helpful

-0.64

uers

-0.64

actionDate

-0.63

ilers

-0.63

Trigger

-0.63

 sshd

-0.62

POSITIVE LOGITS

 underwent

0.92

 endured

0.92

 opted

0.85

 wore

0.84

 struggled

0.83

 enjoys

0.83

 survived

0.82

 joked

0.81

 appeared

0.81

 regretted

0.80

Activations Density 0.193%

famous individuals, specifically celebrities and athletes

terms related to celebrity or notable individuals in various contexts

No Comments

No Known Activations