INDEX

Explanations

terms related to simplicity and ease of use in processes

oai_token-act-pair · gpt-4o-mini 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

lees

-0.95

anwhile

-0.90

aucuses

-0.88

olitics

-0.83

onduct

-0.81

aters

-0.79

Officials

-0.78

politics

-0.78

anches

-0.78

ources

-0.78

POSITIVE LOGITS

 version

1.29

 example

1.19

 tutorial

1.11

 sized

1.06

 solution

1.04

 edition

1.03

 prototype

1.02

 demo

1.01

 copy

0.97

 wrapper

0.96

Activations Density 0.370%

terms related to simplicity and ease of use in processes

No Comments

No Known Activations