INDEX

Explanations

references to important episodes or conclusions in a series

oai_token-act-pair · gpt-3.5-turbo

terms related to television series finales

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

tics

-0.81

urg

-0.76

lying

-0.76

artisan

-0.76

uge

-0.75

Cola

-0.74

olds

-0.74

hist

-0.71

absolute

-0.71

guns

-0.71

POSITIVE LOGITS

 finale

1.07

[|

0.88

 prematurely

0.81

 reel

0.80

 spoiler

0.79

 airs

0.78

 adolesc

0.77

 episode

0.76

APR

0.75

adelphia

0.75

Activations Density 0.012%

references to important episodes or conclusions in a series

terms related to television series finales

No Comments

No Known Activations