INDEX

Explanations

specific locations or places within a building

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

utan

-0.77

ibia

-0.77

bow

-0.73

itudinal

-0.73

UAL

-0.72

merga

-0.71

loads

-0.69

amara

-0.67

wana

-0.67

tarians

-0.66

POSITIVE LOGITS

 courthouse

1.12

 apartment

1.08

 Notting

0.94

 closet

0.94

 hotel

0.90

 mansion

0.90

 apartments

0.90

 CafÃ©

0.89

 courtroom

0.88

 doors

0.88

Activations Density 0.222%

specific locations or places within a building

No Comments

No Known Activations