INDEX

Explanations

keywords related to digital interaction or user engagement

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

peria

-0.72

corn

-0.69

conservancy

-0.68

zn

-0.64

Bav

-0.64

cott

-0.64

GE

-0.63

fork

-0.62

statement

-0.61

ciples

-0.60

POSITIVE LOGITS

ivity

1.17

ively

0.96

ually

0.92

iences

0.91

iveness

0.90

ivating

0.88

ships

0.86

ioned

0.84

ativity

0.84

hips

0.84

Activations Density 0.078%

keywords related to digital interaction or user engagement

No Comments

No Known Activations