INDEX

Explanations

actions related to cleaning or maintaining something

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

ollo

-0.62

ranch

-0.59

 Caption

-0.58

brow

-0.57

ilde

-0.55

Tro

-0.54

 ranked

-0.54

ned

-0.53

 redistributed

-0.53

fly

-0.51

POSITIVE LOGITS

 awhile

1.51

 eternity

1.20

 hours

1.16

 longer

1.16

 decades

1.11

 weeks

1.08

 months

1.08

 millennia

1.08

hours

1.06

months

1.06

Activations Density 1.137%

actions related to cleaning or maintaining something

No Comments

No Known Activations