INDEX

Explanations

mentions of specific locations or entities within a document

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 misunder

-0.64

 amounts

-0.63

 cape

-0.61

 scenery

-0.60

 unfocusedRange

-0.60

 streng

-0.59

 agre

-0.59

icultural

-0.59

 encour

-0.59

 confir

-0.58

POSITIVE LOGITS

ARE

0.72

âĹ¼

0.67

icial

0.66

teenth

0.65

ife

0.64

IFE

0.63

those

0.63

uits

0.62

cheon

0.62

ĳå£«

0.60

Activations Density 0.121%

mentions of specific locations or entities within a document

No Comments

No Known Activations

mentions of specific locations or entities within a document

No Comments

No Known Activations