INDEX

Explanations

references to specific entities being the focal point of discussion

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

/"

-0.60

egu

-0.57

—"

-0.57

 alongside

-0.56

ãĥ¼ãĥĨ

-0.56

 beforehand

-0.56

Ãĥ

-0.55

Keefe

-0.55

omever

-0.54

—-

-0.54

POSITIVE LOGITS

resa

0.88

odore

0.77

atre

0.75

 greatest

0.72

quickShipAvailable

0.72

 longest

0.71

 largest

0.71

 hottest

0.70

orem

0.69

nce

0.68

Activations Density 0.095%

references to specific entities being the focal point of discussion

No Comments

No Known Activations