INDEX

Explanations

references to specific bills and legislative actions

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 corps

-0.77

arya

-0.75

rower

-0.73

 foss

-0.72

bourg

-0.70

 Carnegie

-0.70

 physi

-0.67

 nerves

-0.65

 havoc

-0.63

 semin

-0.63

POSITIVE LOGITS

 Proposition

1.01

 Amendments

0.92

 legalizing

0.84

armor

0.81

 amended

0.79

 Amendment

0.78

 authorizing

0.75

 repealed

0.74

 repeal

0.74

SB

0.73

Activations Density 0.063%

references to specific bills and legislative actions

No Comments

No Known Activations