INDEX

Explanations

proper nouns or specific phrases identifying individuals or locations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating someone's ability or potential to perform an action

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 hating

-0.68

 striving

-0.63

 chasing

-0.60

 Shaw

-0.60

 stalking

-0.59

 rejecting

-0.58

 rejection

-0.58

 Confeder

-0.58

 Simone

-0.58

bush

-0.56

POSITIVE LOGITS

't

1.35

adian

1.18

berra

1.17

isters

1.08

 afford

1.00

asta

0.96

NOT

0.96

 attest

0.93

ister

0.91

idate

0.89

Activations Density 0.244%

proper nouns or specific phrases identifying individuals or locations

phrases indicating someone's ability or potential to perform an action

No Comments

No Known Activations