INDEX

Explanations

references to characters and elements from fictional or mythological stories, particularly related to princesses or mythological figures

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

 Peninsula

-0.71

 thirds

-0.66

Äĩ

-0.63

agine

-0.62

rost

-0.58

 dear

-0.58

asted

-0.57

 damn

-0.56

 nerv

-0.55

 Harbour

-0.55

POSITIVE LOGITS

es

2.48

ively

1.43

esville

1.26

ions

1.24

ed

1.23

esian

1.19

engers

1.18

esy

1.11

eson

1.09

ES

1.07

Activations Density 0.094%

references to characters and elements from fictional or mythological stories, particularly related to princesses or mythological figures

No Comments

No Known Activations