INDEX

Explanations

mentions of hair styles

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

hyde

-0.84

PORT

-0.73

 Hendricks

-0.68

 Mine

-0.68

 Millennium

-0.67

 Abyss

-0.67

 Dragonbound

-0.66

 Fargo

-0.64

 Master

-0.64

Ô

-0.63

POSITIVE LOGITS

inging

1.11

idd

1.02

inged

1.02

iddling

1.01

akings

0.97

iper

0.96

tro

0.96

ashing

0.95

Ð¾

0.94

udging

0.94

Activations Density 1.072%

mentions of hair styles

No Comments

No Known Activations