INDEX

Explanations

mentions of political figures and events

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

 GOODMAN

-0.81

OUP

-0.72

 Drawn

-0.71

ãĤ¤ãĥĪ

-0.69

 Fever

-0.65

 Sabha

-0.65

 Sawyer

-0.64

 Flavoring

-0.59

MSN

-0.59

arity

-0.59

POSITIVE LOGITS

clamation

1.41

orbit

1.26

uber

1.26

ogenous

1.22

terior

1.14

tern

1.10

clus

1.08

portation

1.07

clud

1.07

oplan

1.07

Activations Density 0.012%

mentions of political figures and events

No Comments

No Known Activations