INDEX

Explanations

mentions of specific names, particularly "McCoy" and "McCann"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "McCoy" and related variations

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

 backward

-0.65

 selective

-0.62

gger

-0.62

 Gemini

-0.60

omaly

-0.60

 Shattered

-0.59

estamp

-0.59

 disappointed

-0.59

 deaf

-0.58

 pissed

-0.58

POSITIVE LOGITS

artney

1.07

inelli

1.06

McC

0.98

 McCoy

0.87

onduct

0.83

keyes

0.82

otti

0.82

 McCorm

0.81

iet

0.80

 Mellon

0.79

Activations Density 0.002%

mentions of specific names, particularly "McCoy" and "McCann"

mentions of the name "McCoy" and related variations

No Comments

No Known Activations