INDEX

Explanations

keywords related to the endings of TV shows or movie seasons

oai_token-act-pair · gpt-3.5-turbo

references to season and series finales in television shows

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

 millennial

-1.11

 millennials

-1.10

 evangelical

-0.96

 Millenn

-0.90

 Millennials

-0.81

 evangelicals

-0.80

Nom

-0.77

 Mormons

-0.74

tm

-0.70

 cues

-0.69

POSITIVE LOGITS

 finale

2.64

bone

1.74

Kou

1.08

 premiere

1.05

bones

1.04

 corps

0.89

 climax

0.88

 Canaver

0.83

 Arena

0.82

anie

0.79

Activations Density 0.024%

keywords related to the endings of TV shows or movie seasons

references to season and series finales in television shows

No Comments

No Known Activations