INDEX

Explanations

information related to product packaging and listings (e.g., names, sizes, prices, descriptions)

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

nesota

-0.84

atio

-0.71

baugh

-0.65

OTO

-0.64

 Hots

-0.62

 Dame

-0.62

alties

-0.61

abama

-0.60

shire

-0.58

ansson

-0.58

POSITIVE LOGITS

 containing

1.00

mates

0.97

gie

0.92

 Contents

0.90

aging

0.89

ainers

0.88

 container

0.88

 containers

0.86

 Crate

0.84

arding

0.84

Activations Density 1.322%

information related to product packaging and listings (e.g., names, sizes, prices, descriptions)

No Comments

No Known Activations