INDEX

Explanations

proper nouns related to locations or businesses, particularly those containing the word "Garden"

oai_token-act-pair · gpt-3.5-turbo

references to "Garden," likely indicating a focus on a specific location or venue

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

esty

-0.78

 derail

-0.76

DonaldTrump

-0.74

merce

-0.72

aundering

-0.71

staking

-0.70

ufact

-0.69

etimes

-0.69

 carbohyd

-0.68

senal

-0.68

POSITIVE LOGITS

 Garden

0.99

hire

0.90

 Grove

0.89

keeper

0.86

arden

0.83

 Bowl

0.82

hill

0.81

esville

0.80

éļ

0.79

Aid

0.75

Activations Density 0.008%

proper nouns related to locations or businesses, particularly those containing the word "Garden"

references to "Garden," likely indicating a focus on a specific location or venue

No Comments

No Known Activations