INDEX

Explanations

I'm sorry, but based on the filtered activations provided for Neuron 4, it is difficult to determine a specific pattern or theme that the neuron is looking for

oai_token-act-pair · gpt-3.5-turbo

references to "Price" in various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 skelet

-0.78

 marsh

-0.75

 bast

-0.70

 hots

-0.69

saf

-0.66

ocy

-0.65

 junction

-0.65

*/(

-0.65

 joint

-0.64

 corpus

-0.63

POSITIVE LOGITS

 Price

3.95

Price

2.75

price

2.20

 price

2.12

 Prices

2.06

 Pricing

1.90

 prices

1.68

 pricing

1.57

 Cost

1.56

 priced

1.46

Activations Density 0.014%

I'm sorry, but based on the filtered activations provided for Neuron 4, it is difficult to determine a specific pattern or theme that the neuron is looking for

references to "Price" in various contexts

No Comments

No Known Activations