INDEX

Explanations

mentions of the upper class or high positions

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

tnc

-0.74

 Machina

-0.74

eln

-0.74

-0.72

ensable

-0.71

printf

-0.70

atile

-0.69

Wikipedia

-0.68

inatory

-0.67

roma

-0.67

POSITIVE LOGITS

most

1.23

 lobe

0.89

class

0.80

 earners

0.80

jaw

0.79

 bounds

0.78

 limb

0.77

lip

0.77

 peninsula

0.76

 tier

0.74

Activations Density 0.024%

mentions of the upper class or high positions

No Comments

No Known Activations