INDEX

Explanations

words related to physical locations and items with some focus on negative aspects

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

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

utical

-0.66

 Flavoring

-0.66

illin

-0.61

iosyncr

-0.58

byter

-0.58

ominated

-0.58

emn

-0.57

compl

-0.57

apego

-0.57

inyl

-0.57

POSITIVE LOGITS

 vicinity

0.90

 drawer

0.85

 freezer

0.81

 attic

0.76

 aisle

0.76

 compartment

0.76

 courtyard

0.76

 closet

0.75

 area

0.74

 cage

0.72

Activations Density 0.281%

words related to physical locations and items with some focus on negative aspects

No Comments

No Known Activations