INDEX

Explanations

words related to importance or relevance

oai_token-act-pair · gpt-3.5-turbo

concepts that indicate importance or significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

guyen

-0.77

ULTS

-0.75

odan

-0.73

ahime

-0.71

ãĥ³ãĤ¸

-0.71

uga

-0.70

©¶æ

-0.70

uthor

-0.69

anium

-0.68

apons

-0.68

POSITIVE LOGITS

 enormously

0.89

 tremendously

0.85

 greatly

0.83

 immensely

0.82

 hugely

0.81

lessly

0.78

 alot

0.76

 insofar

0.72

 less

0.72

 chiefly

0.71

Activations Density 0.044%

words related to importance or relevance

concepts that indicate importance or significance

No Comments

No Known Activations