INDEX

Explanations

mentions of various studios

oai_token-act-pair · gpt-3.5-turbo

references to various film studios

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

 refere

-0.72

Ã±

-0.67

iction

-0.65

hyper

-0.65

ochond

-0.64

nir

-0.61

trigger

-0.61

fully

-0.61

indu

-0.60

hist

-0.60

POSITIVE LOGITS

 Studios

1.28

Studio

0.93

 Workshop

0.92

 Studio

0.92

hesda

0.89

 studios

0.83

 studio

0.81

 Associates

0.80

 Norn

0.78

Works

0.76

Activations Density 0.006%

mentions of various studios

references to various film studios

No Comments

No Known Activations