INDEX

Explanations

words related to healthcare and medical procedures

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

ï¿½ï¿½ï¿½ï¿½

-0.75

PLA

-0.74

DIT

-0.74

 Limits

-0.67

\<

-0.66

[/

-0.64

å§«

-0.63

MAP

-0.63

him

-0.62

 VIDEOS

-0.62

POSITIVE LOGITS

pired

0.96

 glamorous

0.91

 flashy

0.91

imilar

0.91

far

0.89

 much

0.89

 advertised

0.85

 prevalent

0.83

 easily

0.83

 well

0.82

Activations Density 0.029%

words related to healthcare and medical procedures

No Comments

No Known Activations