INDEX

Explanations

proper nouns of people or places

oai_token-act-pair · gpt-3.5-turbo

references to specific people or entities

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

ĪĴ

-1.12

aido

-0.85

ogun

-0.79

abama

-0.79

ħĭ

-0.77

DCS

-0.77

Cub

-0.74

fb

-0.73

oÄŁ

-0.72

¬¼

-0.72

POSITIVE LOGITS

Le

2.64

Le

2.39

le

2.12

LE

1.73

Lep

1.51

Leh

1.49

 Leopard

1.49

le

1.42

Lem

1.38

 Leigh

1.36

Activations Density 0.217%

proper nouns of people or places

references to specific people or entities

No Comments

No Known Activations