INDEX

Explanations

references to examining or taking a detailed look at something

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

printf

-0.69

 tragically

-0.62

[*

-0.59

iatric

-0.58

 owing

-0.57

/**

-0.56

andowski

-0.55

usive

-0.55

heres

-0.55

 respectively

-0.55

POSITIVE LOGITS

 Yourself

1.06

 yourselves

1.01

 yourself

0.90

!:

0.79

cknow

0.75

 Skies

0.73

 Your

0.68

 Pricing

0.68

 thou

0.68

 Favor

0.68

Activations Density 0.236%

references to examining or taking a detailed look at something

No Comments

No Known Activations