INDEX

Explanations

commands and steps in a technical tutorial

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

Vin

-0.65

 Gale

-0.63

 independence

-0.63

 Rolls

-0.62

 careers

-0.62

 Rockefeller

-0.62

 Plaint

-0.61

JFK

-0.60

Yon

-0.60

 Jarrett

-0.59

POSITIVE LOGITS

'm

1.46

've

1.29

'll

1.08

am

1.02

 intend

1.00

 presume

0.98

EEE

0.98

WI

0.97

 recommend

0.97

'd

0.96

Activations Density 0.261%

commands and steps in a technical tutorial

No Comments

No Known Activations