INDEX

Explanations

references to actors and their specific roles in films or TV shows

oai_token-act-pair · gpt-3.5-turbo

references to roles and performances in film and television

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

£ı

-0.91

pex

-0.76

 newsletters

-0.76

foundation

-0.70

 redist

-0.68

 caches

-0.68

 Publishers

-0.68

lag

-0.67

 promulg

-0.65

 Pastebin

-0.65

POSITIVE LOGITS

 starring

1.09

 roles

1.05

 actress

1.01

acters

0.97

 portraying

0.93

 actor

0.92

Alias

0.90

 Natalie

0.89

 opposite

0.88

 role

0.86

Activations Density 0.113%

references to actors and their specific roles in films or TV shows

references to roles and performances in film and television

No Comments

No Known Activations