INDEX

Explanations

positive attributes or behaviors related to social responsibility and integrity

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

abet

-0.77

aters

-0.73

lance

-0.73

weeney

-0.73

adata

-0.71

Ĥ¬

-0.69

redits

-0.69

atus

-0.68

imum

-0.68

POSITIVE LOGITS

 alike

1.16

 respectively

0.80

 sounding

0.73

 nature

0.72

 manner

0.69

 striped

0.68

 ways

0.68

 personality

0.67

 combination

0.67

 demeanor

0.63

Activations Density 0.157%

positive attributes or behaviors related to social responsibility and integrity

No Comments

No Known Activations