INDEX

Explanations

the word "David" in various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

cffffcc

-0.98

NetMessage

-0.86

ĺħ

-0.79

phies

-0.78

vana

-0.74

Reply

-0.73

ĻĤ

-0.73

inventoryQuantity

-0.72

ttes

-0.71

¥ŀ

-0.71

POSITIVE LOGITS

 Bowie

1.05

 Cameron

0.99

 Axel

0.95

 Rockefeller

0.91

 Beckham

0.90

Ic

0.90

 Petraeus

0.90

son

0.88

 Barton

0.87

 Misc

0.85

Activations Density 0.015%

the word "David" in various contexts

No Comments

No Known Activations