INDEX

Explanations

references or descriptions related to unnamed or unidentified entities

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

Negative Logits

enegger

-0.80

akings

-0.79

arium

-0.77

âķĲâķĲ

-0.76

lov

-0.74

blems

-0.73

nea

-0.72

âķĲ

-0.72

byss

-0.71

urgy

-0.70

POSITIVE LOGITS

 narrator

1.12

 unnamed

1.04

 assailants

0.99

 persons

0.98

 assailant

0.94

 bystand

0.93

 person

0.91

 informant

0.86

 anonymous

0.85

 unidentified

0.84

Activations Density 0.045%

references or descriptions related to unnamed or unidentified entities

No Known Activations