INDEX

Explanations

legal and policy-related terms and concepts

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

atomic

-0.88

 livest

-0.80

omatic

-0.72

efully

-0.71

 Remem

-0.67

erd

-0.66

eful

-0.64

 Reincarn

-0.64

plings

-0.62

 Afric

-0.62

POSITIVE LOGITS

suit

1.02

suits

1.01

orneys

0.89

 fees

0.87

yer

0.85

 attorney

0.83

 filings

0.82

yers

0.82

 Gloria

0.77

 solicitor

0.76

Activations Density 1.484%

legal and policy-related terms and concepts

No Comments

No Known Activations