INDEX

Explanations

phrases related to legal terms, regulations, and processes

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

sth

-0.79

Reviewer

-0.72

worldly

-0.66

 Dragonbound

-0.63

 Boll

-0.63

)=(

-0.63

ï¸

-0.62

 Holly

-0.61

 Elves

-0.61

DAY

-0.60

POSITIVE LOGITS

initions

1.35

ensible

1.26

endants

1.26

initely

1.23

endant

1.21

ensions

1.15

erence

1.14

icient

1.14

ibr

1.13

erred

1.12

Activations Density 0.498%

phrases related to legal terms, regulations, and processes

No Comments

No Known Activations