INDEX

Explanations

references to significant events or issues related to warfare or conflict

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 brunt

-0.69

 claws

-0.69

rou

-0.66

 tear

-0.66

 rolls

-0.66

gre

-0.65

 benches

-0.64

hement

-0.64

 chops

-0.64

 nails

-0.64

POSITIVE LOGITS

Overview

1.33

Introduction

1.12

Â¶

1.07

Understanding

1.06

ccording

1.02

Posted

1.02

Definition

0.97

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

0.97

Summary

0.95

Synopsis

0.94

Activations Density 0.137%

references to significant events or issues related to warfare or conflict

No Comments

No Known Activations