INDEX

Explanations

phrases related to popular movies, TV shows, and characters

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

STEM

-0.70

ILLE

-0.66

ONY

-0.66

uliffe

-0.65

arette

-0.64

 Nieto

-0.63

nown

-0.60

mpeg

-0.59

*/(

-0.59

agents

-0.59

POSITIVE LOGITS

 Admir

0.82

 Rings

0.77

warts

0.69

 Apostles

0.66

 Thieves

0.65

swer

0.65

 Prayer

0.64

 Canterbury

0.62

 Sheep

0.61

 Scrolls

0.60

Activations Density 0.042%

phrases related to popular movies, TV shows, and characters

No Comments

No Known Activations