INDEX

Explanations

terms related to specific people or events, likely related to a historical context

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 damned

-0.79

 damn

-0.72

 crunch

-0.71

 shoppers

-0.71

 darn

-0.70

 PLUS

-0.69

utterstock

-0.68

 guessing

-0.67

thouse

-0.66

(@

-0.66

POSITIVE LOGITS

Appearances

1.39

Contents

1.34

References

1.31

Trivia

1.29

tymology

1.18

Notable

1.15

Gallery

1.11

History

1.10

edit

1.04

Notes

1.04

Activations Density 2.898%

terms related to specific people or events, likely related to a historical context

No Comments

No Known Activations