INDEX

Explanations

percentages in relation to political matchups

oai_token-act-pair · gpt-3.5-turbo

phrases indicating comparative results or outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

calling

-0.61

é¾įåĸļå£«

-0.60

 largeDownload

-0.59

Rew

-0.59

 Donation

-0.58

hop

-0.57

cffffcc

-0.57

itudinal

-0.56

 exceptions

-0.55

 evid

-0.54

POSITIVE LOGITS

ggles

1.11

 date

0.99

 qualify

0.86

 compensate

0.83

ppers

0.80

asted

0.80

date

0.78

wered

0.76

adies

0.73

 settle

0.73

Activations Density 0.087%

percentages in relation to political matchups

phrases indicating comparative results or outcomes

No Comments

No Known Activations