INDEX

Explanations

information indicating things to take note of, be mindful of, or know about

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing important points or guidelines to consider

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

 Cros

-0.70

netflix

-0.63

Publisher

-0.62

reality

-0.60

rek

-0.60

tur

-0.59

 corp

-0.57

unic

-0.57

geon

-0.56

 Tens

-0.56

POSITIVE LOGITS

 happen

1.04

 notice

1.00

 consider

0.97

 beware

0.84

 worry

0.83

 note

0.83

 mention

0.82

 Notice

0.81

 remember

0.81

 discuss

0.81

Activations Density 0.150%

information indicating things to take note of, be mindful of, or know about

phrases emphasizing important points or guidelines to consider

No Comments

No Known Activations