INDEX

Explanations

phrases related to authority and control

oai_token-act-pair · gpt-3.5-turbo

elements related to addiction and dependency issues

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

qua

-0.72

 Peaks

-0.68

paralle

-0.67

qa

-0.63

 Press

-0.61

ebus

-0.60

echo

-0.58

 Podcast

-0.58

framework

-0.58

 Tyrann

-0.57

POSITIVE LOGITS

 themselves

1.72

 their

1.22

 THEIR

1.08

 selves

1.04

their

1.03

 careers

1.01

 voluntarily

0.98

utterstock

0.92

selves

0.89

Their

0.88

Activations Density 0.840%

phrases related to authority and control

elements related to addiction and dependency issues

No Comments

No Known Activations