INDEX

Explanations

film director names

oai_token-act-pair · gpt-3.5-turbo

references to film directors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Peb

-0.79

tics

-0.70

achine

-0.66

 Canadians

-0.65

 Ukrainians

-0.63

 Blueprint

-0.62

Ukraine

-0.61

 Kimmel

-0.61

orthy

-0.59

orum

-0.59

POSITIVE LOGITS

igible

1.14

Dir

0.84

der

0.81

ried

0.80

iger

0.78

vana

0.78

idon

0.76

nit

0.75

icular

0.75

dir

0.75

Activations Density 0.010%

film director names

references to film directors

No Comments

No Known Activations