INDEX

Explanations

sentences with information about medical conditions, treatments, recoveries, and medical facilities

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

interstitial

-0.71

igious

-0.69

uca

-0.68

 Inferno

-0.68

 disproportion

-0.67

 collide

-0.66

ire

-0.64

 inappropriate

-0.63

 incest

-0.63

erv

-0.62

POSITIVE LOGITS

 satisfactory

0.97

 remission

0.91

 stabilized

0.91

 unchanged

0.90

 restored

0.90

 congratulations

0.89

 Congratulations

0.84

 congr

0.83

 improved

0.83

albeit

0.83

Activations Density 0.904%

sentences with information about medical conditions, treatments, recoveries, and medical facilities

No Comments

No Known Activations