INDEX

Explanations

terms related to testing, reviews, and evaluations of software and technology products

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

taboola

-0.74

ACP

-0.70

resent

-0.70

theless

-0.68

vernment

-0.66

 traged

-0.64

ignt

-0.62

wikipedia

-0.62

lance

-0.61

 appropriations

-0.61

POSITIVE LOGITS

osterone

1.43

imony

1.21

udo

1.07

imon

1.00

icle

0.87

icles

0.87

 Testing

0.83

 dummy

0.80

 test

0.79

 testing

0.78

Activations Density 1.937%

terms related to testing, reviews, and evaluations of software and technology products

No Comments

No Known Activations