INDEX

Explanations

words related to recognition or awareness

oai_token-act-pair · gpt-3.5-turbo

words related to recognition and cognizance

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

bury

-0.72

 mileage

-0.68

scenes

-0.67

hours

-0.66

Far

-0.65

 Beir

-0.65

RESULTS

-0.62

-0.61

 setbacks

-0.61

Mek

-0.61

POSITIVE LOGITS

isance

1.10

ocent

1.09

ition

1.09

icious

1.03

itives

1.02

ational

1.00

omore

0.99

ogn

0.98

ormal

0.97

ificent

0.97

Activations Density 0.007%

words related to recognition or awareness

words related to recognition and cognizance

No Comments

No Known Activations