INDEX

Explanations

references to the popular franchise "Pokémon."

oai_token-act-pair · gpt-3.5-turbo

mentions of the Pokémon franchise

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

iffe

-0.83

ndra

-0.72

perial

-0.72

outer

-0.72

ubb

-0.71

ijk

-0.69

raught

-0.68

lishes

-0.68

inges

-0.68

ickson

-0.67

POSITIVE LOGITS

Dex

1.00

PokÃ©mon

0.92

 pokemon

0.91

 PokÃ©mon

0.91

 Pokemon

0.88

athlon

0.85

 Trainer

0.82

 PokÃ©

0.82

Gy

0.80

 Pikachu

0.79

Activations Density 0.012%

references to the popular franchise "Pokémon."

mentions of the Pokémon franchise

No Comments

No Known Activations