INDEX

Explanations

comparisons indicating an increase or enhancement

oai_token-act-pair · gpt-3.5-turbo

a high frequency of the word "more."

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

xtap

-0.73

Enlarge

-0.72

 Classification

-0.65

-0.64

MIT

-0.64

 Films

-0.63

OCK

-0.63

oresc

-0.62

DragonMagazine

-0.62

POSITIVE LOGITS

 than

1.22

 stringent

0.99

 expensive

0.94

 sophisticated

0.93

ado

0.91

 efficient

0.89

 importantly

0.89

 complicated

0.88

 elaborate

0.87

 costly

0.87

Activations Density 0.167%

comparisons indicating an increase or enhancement

a high frequency of the word "more."

No Comments

No Known Activations