INDEX

Explanations

mentions of individuals engaging in activities or actions

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

OTS

-0.78

UTF

-0.69

Â¯

-0.68

ELF

-0.62

acters

-0.60

cpp

-0.59

 âī¡

-0.59

utf

-0.58

YP

-0.58

POSITIVE LOGITS

 hiatus

0.99

 sale

0.90

stage

0.84

shore

0.82

ibaba

0.81

 autop

0.80

boarding

0.80

Downloadha

0.79

 board

0.78

 display

0.78

Activations Density 0.060%

mentions of individuals engaging in activities or actions

No Comments

No Known Activations