INDEX

Explanations

locations or establishments

oai_token-act-pair · gpt-3.5-turbo

words related to professions, particularly those ending in 'iner'

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

hovah

-0.92

enegger

-0.83

akening

-0.77

soDeliveryDate

-0.71

sburgh

-0.71

stretched

-0.67

ecause

-0.66

isSpecialOrderable

-0.64

painted

-0.64

sheets

-0.64

POSITIVE LOGITS

iner

0.87

iners

0.81

idia

0.77

oco

0.76

baum

0.75

cest

0.75

izontal

0.74

istics

0.73

nesota

0.73

TAIN

0.73

Activations Density 0.024%

locations or establishments

words related to professions, particularly those ending in 'iner'

No Comments

No Known Activations