INDEX

Explanations

words related to the Pokémon franchise

oai_token-act-pair · gpt-3.5-turbo

references to Pokémon and related concepts

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

)=(

-0.75

ufact

-0.71

yer

-0.66

cens

-0.65

 offic

-0.65

lishes

-0.63

 McDonnell

-0.63

perial

-0.62

×Ļ

-0.62

lain

-0.62

POSITIVE LOGITS

okemon

0.96

Dex

0.94

Pokemon

0.92

 pokemon

0.91

 Pokemon

0.86

Gy

0.80

athlon

0.79

XY

0.79

phan

0.77

walker

0.75

Activations Density 0.016%

words related to the Pokémon franchise

references to Pokémon and related concepts

No Comments

No Known Activations