INDEX

Explanations

phrases related to success or accomplishment

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anwhile

-0.87

 Monitor

-0.86

esa

-0.76

emporary

-0.72

Monitor

-0.71

orters

-0.70

oulder

-0.69

arine

-0.68

awn

-0.68

ynes

-0.68

POSITIVE LOGITS

 successfully

0.86

 miraculous

0.85

 spectacular

0.84

 coup

0.83

 victory

0.82

 stunts

0.81

 tricks

0.81

 unexpected

0.81

 miracle

0.79

 deception

0.78

Activations Density 0.037%

phrases related to success or accomplishment

No Comments

No Known Activations