INDEX

Explanations

phrases related to watching or reading content

oai_token-act-pair · gpt-3.5-turbo

references to video content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

anc

-0.80

establishment

-0.69

omorphic

-0.68

cedented

-0.68

roots

-0.68

©¶æ¥µ

-0.66

 unaccount

-0.66

/(

-0.65

 oxid

-0.64

poons

-0.64

POSITIVE LOGITS

 VIDEOS

0.94

 unfold

0.83

 clip

0.76

 HERE

0.74

 replay

0.70

 Hulu

0.69

 yours

0.69

 CLICK

0.68

||||

0.68

 Scrib

0.68

Activations Density 0.364%

phrases related to watching or reading content

references to video content

No Comments

No Known Activations