INDEX

Explanations

words related to medical conditions, such as diagnosis and cancer

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

urch

-0.77

©¶æ¥µ

-0.77

obile

-0.69

oca

-0.69

burse

-0.68

Lago

-0.68

gage

-0.66

bris

-0.66

usc

-0.65

ibi

-0.65

POSITIVE LOGITS

 misconceptions

1.07

 pitfalls

1.02

 discusses

1.01

 insights

1.00

 tips

0.98

 implications

0.95

 topics

0.95

 Lessons

0.94

 lessons

0.94

 hints

0.94

Activations Density 0.370%

words related to medical conditions, such as diagnosis and cancer

No Comments

No Known Activations