INDEX

Explanations

words related to consuming beverages, specifically the act of drinking

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of the word "drinking," particularly in relation to beverages and consumption contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Sharif

-0.72

eele

-0.71

ural

-0.69

locked

-0.68

ominated

-0.67

merce

-0.64

pora

-0.63

 Decay

-0.62

moving

-0.62

theless

-0.62

POSITIVE LOGITS

cohol

1.04

 alcohol

1.04

 bottles

1.00

 drinking

0.99

 water

0.96

water

0.95

 gallons

0.95

 beverages

0.91

 soda

0.91

 bleach

0.91

Activations Density 0.022%

words related to consuming beverages, specifically the act of drinking

repeated mentions of the word "drinking," particularly in relation to beverages and consumption contexts

No Comments

No Known Activations