INDEX

Explanations

references to companies or entities associated with the word "Pictures."

oai_token-act-pair · gpt-3.5-turbo

mentions of film production companies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 emer

-0.78

 increment

-0.68

leg

-0.65

 element

-0.65

 obligation

-0.65

 complement

-0.65

abled

-0.63

 healing

-0.62

 sustain

-0.62

 transitioning

-0.62

POSITIVE LOGITS

 Pictures

4.14

pict

2.12

 Picture

1.93

Pict

1.90

 pictures

1.72

 Images

1.66

 Photos

1.66

 Films

1.63

 pics

1.45

photos

1.44

Activations Density 0.007%

references to companies or entities associated with the word "Pictures."

mentions of film production companies

No Comments

No Known Activations