INDEX

Explanations

references to cinema or related activities

oai_token-act-pair · gpt-3.5-turbo

references to cinema or films in various contexts

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

legged

-0.69

lynn

-0.69

lies

-0.68

rab

-0.66

lying

-0.66

ciating

-0.65

pages

-0.63

hee

-0.62

nesses

-0.61

egg

-0.61

POSITIVE LOGITS

 cinema

1.19

 Cinema

1.08

 theaters

0.98

 cinem

0.94

goers

0.91

 theatre

0.90

 theat

0.88

netflix

0.88

 kios

0.87

 marqu

0.85

Activations Density 0.007%

references to cinema or related activities

references to cinema or films in various contexts

No Comments

No Known Activations