INDEX

Explanations

proper nouns related to a specific individual named Thor

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

 sterling

-0.71

pring

-0.70

VICE

-0.67

URES

-0.67

aminer

-0.67

holders

-0.65

 VIDEOS

-0.63

FTA

-0.62

Keefe

-0.62

pace

-0.62

POSITIVE LOGITS

 Ragnarok

1.19

sten

1.16

nton

1.15

aldi

0.96

bj

0.93

grim

0.92

azine

0.91

agnar

0.88

stein

0.86

Thor

0.84

Activations Density 0.016%

proper nouns related to a specific individual named Thor

No Comments

No Known Activations