INDEX

Explanations

topics related to politics, regulations, and lobbying

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

DOS

-0.76

 Sons

-0.71

guyen

-0.69

amaz

-0.68

CHAT

-0.67

owicz

-0.66

ramids

-0.64

 teasp

-0.64

anny

-0.64

Enough

-0.64

POSITIVE LOGITS

 imaginable

1.26

 including

1.20

viz

1.04

 namely

1.01

 affecting

1.00

 besides

0.96

 pertaining

0.95

 relating

0.95

 ranging

0.94

 such

0.90

Activations Density 0.595%

topics related to politics, regulations, and lobbying

No Comments

No Known Activations

topics related to politics, regulations, and lobbying

No Comments

No Known Activations