INDEX

Explanations

personal exclamations or declarations

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

 Mormonism

-0.66

 Constitutional

-0.64

 Failure

-0.64

 Ezek

-0.63

 Uganda

-0.60

 Permanent

-0.60

 Gund

-0.60

 Portal

-0.59

 Grateful

-0.59

 Eternity

-0.58

POSITIVE LOGITS

 shouldn

1.04

'll

1.02

ctic

0.92

 gotta

0.91

 dunno

0.89

 naturally

0.87

pher

0.85

're

0.85

 gladly

0.84

ems

0.84

Activations Density 0.131%

personal exclamations or declarations

No Comments

No Known Activations