INDEX

Explanations

phrases describing events or states that occurred or existed in the past

oai_token-act-pair · gpt-3.5-turbo

phrases indicating historical references or events that occurred before a specific time

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ãĤĬ

-0.74

gallery

-0.73

Pool

-0.69

egu

-0.69

ihar

-0.69

erves

-0.69

<?

-0.69

cipl

-0.69

urden

-0.67

arse

-0.67

POSITIVE LOGITS

 advent

1.13

 Gutenberg

1.00

 WWII

0.96

 globalization

0.95

0.94

 puberty

0.94

 modern

0.93

 Prohibition

0.91

 invention

0.87

 Enlightenment

0.87

Activations Density 0.208%

phrases describing events or states that occurred or existed in the past

phrases indicating historical references or events that occurred before a specific time

No Comments

No Known Activations