INDEX

Explanations

phrases or sentences indicating a specific position or status of something

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

nces

-0.77

................................................................

-0.72

nick

-0.71

aic

-0.70

incent

-0.70

aeus

-0.70

eal

-0.65

oaded

-0.65

ln

-0.63

ixt

-0.63

POSITIVE LOGITS

 society

1.11

 hierarchy

0.87

 Parliament

0.86

 parliament

0.86

ADRA

0.79

 UNESCO

0.78

 prestigious

0.77

 elite

0.76

 academia

0.76

thood

0.73

Activations Density 0.223%

phrases or sentences indicating a specific position or status of something

No Comments

No Known Activations