INDEX

Explanations

words related to alcoholic beverages, especially whiskey and bourbon

oai_token-act-pair · gpt-3.5-turbo

references to various types of whiskey and bourbon

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Parenthood

-0.81

ularity

-0.81

lez

-0.74

ULAR

-0.73

NCT

-0.73

 Submission

-0.71

ospace

-0.68

undai

-0.66

 Telecommunications

-0.66

AFTA

-0.64

POSITIVE LOGITS

 whiskey

1.38

 whisky

1.25

 Whis

1.10

illery

0.99

 bourbon

0.99

wine

0.98

 distilled

0.95

 vinegar

0.93

 liquor

0.92

 vodka

0.91

Activations Density 0.023%

words related to alcoholic beverages, especially whiskey and bourbon

references to various types of whiskey and bourbon

No Comments

No Known Activations