INDEX

Explanations

phrases related to criticism and assessment

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

âĢ¦)

-0.58

)"

-0.57

...)

-0.56

Loading

-0.54

inea

-0.53

!)

-0.53

 Dise

-0.52

avia

-0.51

 gener

-0.50

 physic

-0.50

POSITIVE LOGITS

etheless

1.07

iannopoulos

0.62

 Pruitt

0.60

 Trotsky

0.60

 Gorsuch

0.58

AFP

0.57

 âĵĺ

0.56

 Stalin

0.53

climate

0.53

orbit

0.52

Activations Density 2.489%

phrases related to criticism and assessment

No Comments

No Known Activations