INDEX

Explanations

words related to toys

oai_token-act-pair · gpt-3.5-turbo

references to toys

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Asheville

-0.77

 Teresa

-0.76

osta

-0.73

ences

-0.72

GP

-0.69

pperc

-0.68

hens

-0.68

 Athens

-0.66

ALS

-0.66

OSE

-0.66

POSITIVE LOGITS

toy

3.55

 toys

3.01

Toy

2.40

Toy

2.24

 Toys

2.24

 dolls

1.74

 doll

1.64

 Lego

1.48

 gadget

1.36

 Barbie

1.32

Activations Density 0.019%

words related to toys

references to toys

No Comments

No Known Activations