INDEX

Explanations

specific names (e.g. "Mary", "Chen") and mention of events or dates

oai_token-act-pair · gpt-3.5-turbo

references to legal or justice-related issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ãĤ´ãĥ³

-0.96

 Odyssey

-0.87

Rei

-0.86

LCS

-0.85

 Apex

-0.84

 NASL

-0.84

 lighthouse

-0.81

 Lilith

-0.80

imum

-0.77

imer

-0.76

POSITIVE LOGITS

Che

2.46

Che

2.27

Chel

1.84

che

1.81

che

1.80

 Cheng

1.29

CH

1.26

 Chel

1.23

 chees

1.18

 Chev

1.16

Activations Density 0.213%

specific names (e.g. "Mary", "Chen") and mention of events or dates

references to legal or justice-related issues

No Comments

No Known Activations