INDEX

Explanations

anomalies or abnormalities within various contexts

oai_token-act-pair · gpt-3.5-turbo

terms related to anomalies and abnormal conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

rol

-1.03

sonian

-1.00

neys

-0.90

ney

-0.83

ystem

-0.83

ofi

-0.81

ulf

-0.79

lasses

-0.79

uay

-0.79

han

-0.79

POSITIVE LOGITS

 anomaly

1.17

 anomalies

1.07

oscope

0.87

ously

0.77

querade

0.72

 Detected

0.72

 anomal

0.67

 detection

0.67

artifacts

0.66

 flare

0.66

Activations Density 0.022%

anomalies or abnormalities within various contexts

terms related to anomalies and abnormal conditions

No Comments

No Known Activations