INDEX

Explanations

proper nouns associated with locations and people

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

terday

-0.69

 expires

-0.66

wip

-0.65

 Adin

-0.63

 laureate

-0.62

 contr

-0.60

 diapers

-0.60

 bottleneck

-0.59

TTL

-0.59

cipl

-0.59

POSITIVE LOGITS

orea

1.15

EEP

1.06

rieg

1.05

irk

1.05

essel

1.04

laus

1.04

ratom

1.01

eeper

1.00

won

0.99

ernel

0.99

Activations Density 0.047%

proper nouns associated with locations and people

No Comments

No Known Activations