INDEX

Explanations

words related to anomalies or abnormal situations

oai_token-act-pair · gpt-3.5-turbo

terms related to anomalies and abnormality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

-1.00

ney

-0.88

uay

-0.85

nan

-0.84

shaw

-0.80

sol

-0.79

neys

-0.78

amins

-0.77

nam

-0.76

nie

-0.74

POSITIVE LOGITS

 anomaly

0.99

 anomalies

0.93

ously

0.83

 Detected

0.82

isite

0.72

istically

0.67

 occurrence

0.65

oscope

0.65

 occurrences

0.64

 anomal

0.64

Activations Density 0.010%

words related to anomalies or abnormal situations

terms related to anomalies and abnormality

No Comments

No Known Activations