INDEX

Explanations

the word "explosion"

oai_token-act-pair · gpt-3.5-turbo

mentions of explosions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

hist

-0.76

secut

-0.75

stra

-0.74

tern

-0.72

nda

-0.70

ĻĤ

-0.69

Ĥ

-0.68

stood

-0.66

guyen

-0.65

posted

-0.64

POSITIVE LOGITS

 explosion

1.20

 explosions

0.89

 Explosion

0.85

 blasts

0.85

 boom

0.84

 bursting

0.83

 fireball

0.83

 explodes

0.82

 burst

0.81

 blast

0.80

Activations Density 0.011%

the word "explosion"

mentions of explosions

No Comments

No Known Activations