INDEX

Explanations

words related to beer

oai_token-act-pair · gpt-3.5-turbo

mentions of beer

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

heed

-0.82

 Telecommunications

-0.81

 Telecom

-0.70

 Lanka

-0.69

urances

-0.68

UTC

-0.64

aido

-0.63

 metic

-0.62

-0.61

angular

-0.59

POSITIVE LOGITS

 brewed

1.43

 drinkers

1.28

 tasting

1.13

 bottles

1.10

 brewing

1.02

 brewers

1.02

 bottle

1.02

 cans

1.00

 Bottle

0.99

bott

0.96

Activations Density 0.033%

words related to beer

mentions of beer

No Comments

No Known Activations