INDEX

Explanations

statements or questions that indicate critical analysis or examination

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

anchester

-0.72

uly

-0.68

alian

-0.67

 accustomed

-0.66

ctic

-0.65

 dexter

-0.64

fal

-0.64

orts

-0.64

inki

-0.63

thal

-0.63

POSITIVE LOGITS

 namely

1.40

 Revenge

0.77

 Capitalism

0.67

 Return

0.63

AIDS

0.62

 Marijuana

0.61

INS

0.61

 Something

0.61

 Catholicism

0.60

 Lifetime

0.60

Activations Density 0.233%

statements or questions that indicate critical analysis or examination

No Comments

No Known Activations