INDEX

Explanations

names of specific individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to people and their activities

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

izens

-0.72

cyt

-0.67

ntil

-0.67

atform

-0.66

 polar

-0.62

ccording

-0.62

 tourism

-0.60

rolog

-0.59

 discrep

-0.59

ournal

-0.58

POSITIVE LOGITS

Jr

1.33

Sr

1.01

wine

1.01

QC

0.99

baugh

0.96

III

0.96

meier

0.87

baum

0.87

witz

0.86

burgh

0.85

Activations Density 0.237%

names of specific individuals

proper nouns related to people and their activities

No Comments

No Known Activations