INDEX

Explanations

terms related to voting and electronic devices

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

 Undead

-0.73

 guiActiveUnfocused

-0.70

 Johann

-0.66

 Trin

-0.63

 Loren

-0.61

,,,,

-0.60

 Kingdoms

-0.59

 Imper

-0.56

 Franz

-0.56

 Maiden

-0.56

POSITIVE LOGITS

 blockers

0.75

 boarded

0.74

ploy

0.72

 vaccination

0.71

blocking

0.71

 screenings

0.71

 screening

0.69

boarding

0.69

abuse

0.67

 policing

0.67

Activations Density 0.249%

terms related to voting and electronic devices

No Comments

No Known Activations