INDEX

Explanations

words related to the name "Fe" or possibly "Fez"

oai_token-act-pair · gpt-3.5-turbo

references to the element "Fe" or related scientific terms

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

the syllable "Fe"

oai_token-act-pair · gpt-4-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Carbuncle

-0.83

worthy

-0.67

 Overs

-0.67

DOWN

-0.66

appropriately

-0.66

 eleph

-0.64

 Zoom

-0.63

 Protector

-0.63

 ANGEL

-0.63

POSITIVE LOGITS

plin

0.95

uth

0.92

Fe

0.90

esville

0.90

ulner

0.89

vered

0.88

cking

0.88

uer

0.86

Fe

0.86

acebook

0.85

Activations Density 0.007%

words related to the name "Fe" or possibly "Fez"

references to the element "Fe" or related scientific terms

the syllable "Fe"

No Comments

No Known Activations