INDEX

Explanations

references to specific episodes

oai_token-act-pair · gpt-3.5-turbo

mentions of specific episodes of television shows

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ntil

-0.90

enei

-0.83

sheets

-0.78

erness

-0.75

ple

-0.75

zza

-0.74

punk

-0.74

achev

-0.74

wegian

-0.73

lli

-0.73

POSITIVE LOGITS

 episode

0.99

 finale

0.89

 episodes

0.89

 aired

0.87

 airing

0.87

 airs

0.86

 opener

0.82

 Transcript

0.80

 Episode

0.80

Episode

0.79

Activations Density 0.019%

references to specific episodes

mentions of specific episodes of television shows

No Comments

No Known Activations