INDEX

Explanations

keywords related to medical and health topics, potentially focusing on specific companies and practices

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

-0.77

interstitial

-0.74

KER

-0.64

////

-0.63

abilities

-0.62

++++++++++++++++

-0.60

 Predators

-0.60

hips

-0.60

ptions

-0.59

ANS

-0.59

POSITIVE LOGITS

heartedly

1.75

hearted

1.32

meal

1.08

 Foods

0.79

 genome

0.76

hog

0.75

 swat

0.73

allo

0.72

cloth

0.72

 wheat

0.72

Activations Density 0.021%

keywords related to medical and health topics, potentially focusing on specific companies and practices

No Comments

No Known Activations