INDEX

Explanations

units of measure followed by values

oai_token-act-pair · gpt-3.5-turbo

measurements and units specifying quantities in scientific contexts

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

 amounts

-0.85

 bonuses

-0.80

 reminders

-0.73

 unmist

-0.73

 clues

-0.71

 rewards

-0.70

 interchange

-0.70

 Ober

-0.70

 sooner

-0.69

 companion

-0.69

POSITIVE LOGITS

dL

1.30

hr

1.23

each

1.23

sq

1.22

1.16

ea

1.16

hour

1.15

lbs

1.15

sec

1.15

wk

1.14

Activations Density 0.033%

units of measure followed by values

measurements and units specifying quantities in scientific contexts

No Comments

No Known Activations