INDEX

Explanations

health-related terms and medical assessments

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

 Petersen

-0.74

ahime

-0.74

ARA

-0.73

agents

-0.73

enhagen

-0.71

assembly

-0.71

encers

-0.70

 Publications

-0.70

ascript

-0.69

encer

-0.69

POSITIVE LOGITS

bad

1.18

 mediocre

1.12

 weak

1.12

 lousy

1.10

 rotten

1.10

 GOOD

1.10

 terrible

1.05

 good

1.05

 perfect

1.02

 greatness

1.02

Activations Density 0.328%

health-related terms and medical assessments

No Comments

No Known Activations