INDEX

Explanations

phrases related to behind-the-scenes activities

oai_token-act-pair · gpt-3.5-turbo

references to behind-the-scenes content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Polaris

-0.68

fulness

-0.67

 antidepressants

-0.62

 Kaufman

-0.61

 pound

-0.61

FUL

-0.61

 Aristotle

-0.60

sted

-0.60

fully

-0.60

 Latvia

-0.59

POSITIVE LOGITS

scenes

1.28

acters

0.80

scene

0.77

Dialogue

0.75

Mods

0.71

 Scenes

0.70

anie

0.69

EngineDebug

0.69

views

0.69

 Procedures

0.68

Activations Density 0.008%

phrases related to behind-the-scenes activities

references to behind-the-scenes content

No Comments

No Known Activations