INDEX

Explanations

proper nouns

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

kok

-0.68

 Bust

-0.67

wrapper

-0.62

ilon

-0.62

 Remastered

-0.60

 Maze

-0.58

 ASAP

-0.58

DNA

-0.57

 Terrorism

-0.57

 Assessment

-0.56

POSITIVE LOGITS

est

1.04

ever

0.91

eenth

0.90

er

0.88

 mortals

0.86

eric

0.82

eties

0.82

azaki

0.80

een

0.77

 exceptions

0.76

Activations Density 0.065%

proper nouns

No Comments

No Known Activations