INDEX

Explanations

phrases related to defining or redefining concepts

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

athi

-0.79

rentice

-0.74

rites

-0.73

ience

-0.70

riter

-0.68

wash

-0.68

asin

-0.66

ili

-0.66

FactoryReloaded

-0.65

obe

-0.65

POSITIVE LOGITS

 definitions

0.88

 boundaries

0.86

initions

0.86

 thresholds

0.80

 defines

0.80

 parameters

0.79

 criteria

0.79

 defining

0.75

 defined

0.73

 FORM

0.72

Activations Density 0.029%

phrases related to defining or redefining concepts

No Comments

No Known Activations