INDEX

Explanations

names of people with "worth" in their surname

oai_token-act-pair · gpt-3.5-turbo

the term "worth" in various contexts or references to value

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 deaf

-0.72

BLE

-0.63

 heartbeat

-0.63

PDATE

-0.59

 jerk

-0.59

 HELL

-0.59

 Wikileaks

-0.59

IRO

-0.58

 Alexa

-0.58

 Awakening

-0.56

POSITIVE LOGITS

worth

1.10

sburg

0.99

borough

0.96

sworth

0.95

doms

0.91

entin

0.87

iness

0.86

pole

0.83

sung

0.83

nesses

0.82

Activations Density 0.008%

names of people with "worth" in their surname

the term "worth" in various contexts or references to value

No Comments

No Known Activations