INDEX

Explanations

technology-related terms and phrases

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

 earnest

-0.73

azines

-0.69

ãĥ©ãĥ³

-0.67

gets

-0.66

bur

-0.65

joy

-0.64

 gloom

-0.62

hab

-0.62

hiro

-0.62

iddy

-0.61

POSITIVE LOGITS

ologies

1.19

 whereby

1.10

 devised

1.01

atics

1.00

 pioneered

0.92

atically

0.91

matic

0.89

 implemented

0.86

atic

0.85

 operated

0.84

Activations Density 2.781%

technology-related terms and phrases

No Comments

No Known Activations