INDEX

Explanations

technical terms related to scientific research and studies

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

anooga

-0.95

 sucks

-0.75

 thereof

-0.72

SPONSORED

-0.70

nob

-0.69

 sucked

-0.66

å¥

-0.63

Weapon

-0.63

 sugars

-0.62

 slightest

-0.61

POSITIVE LOGITS

Mub

0.71

 several

0.71

 Naval

0.71

 Lockheed

0.69

 Markus

0.69

 Airbus

0.67

two

0.66

 Operation

0.66

Geo

0.66

 numerous

0.65

Activations Density 0.454%

technical terms related to scientific research and studies

No Comments

No Known Activations