INDEX

Explanations

words related to historical events and figures

oai_token-act-pair · gpt-3.5-turbo

references to significant legal or historical figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Reviewer

-0.91

Ĥ¬

-0.75

uated

-0.72

REDACTED

-0.70

kick

-0.63

favorite

-0.62

ting

-0.62

lifting

-0.60

uate

-0.60

ECTION

-0.60

POSITIVE LOGITS

pton

0.95

 shenan

0.89

perty

0.85

olyn

0.83

igham

0.75

ptions

0.75

 Crom

0.74

arre

0.74

iosyn

0.73

bie

0.69

Activations Density 0.011%

words related to historical events and figures

references to significant legal or historical figures

No Comments

No Known Activations