INDEX

Explanations

government-related entities and legal terms

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Lover

-0.96

 Characters

-0.93

 Disorder

-0.92

 Patterns

-0.92

 Pattern

-0.90

 Dates

-0.89

 Lives

-0.89

 Spaces

-0.87

 Owners

-0.87

 Accountability

-0.87

POSITIVE LOGITS

 chief

0.86

 chiefs

0.85

 commissioners

0.83

 commissioner

0.82

 office

0.81

 council

0.78

 franc

0.78

 headquarters

0.78

 academy

0.78

 primates

0.77

Activations Density 11.972%

government-related entities and legal terms

No Comments

No Known Activations