INDEX

Explanations

words related to postal services

oai_token-act-pair · gpt-3.5-turbo

mentions of postal services or related terms

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Flavoring

-0.83

 Qiao

-0.81

estial

-0.75

ãĤ¤ãĥĪ

-0.73

IBLE

-0.71

ONSORED

-0.69

ãĤ¼ãĤ¦ãĤ¹

-0.69

oice

-0.65

aughs

-0.65

 WATCHED

-0.65

POSITIVE LOGITS

gres

1.37

greSQL

1.23

mortem

1.10

doctoral

1.07

erity

1.00

natal

0.93

modern

0.93

hum

0.91

traumatic

0.89

graduate

0.88

Activations Density 0.025%

words related to postal services

mentions of postal services or related terms

No Comments

No Known Activations