INDEX

Explanations

phrases that indicate formal calls to action or requests for responses

oai_token-act-pair · gpt-4o-mini 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

VERTISEMENT

-0.82

puted

-0.78

amac

-0.70

mol

-0.70

served

-0.68

ccording

-0.64

amic

-0.63

ãĤª

-0.63

 âĶľâĶĢâĶĢ

-0.62

////////////////

-0.62

POSITIVE LOGITS

 policymakers

0.98

 governments

0.88

 lawmakers

0.87

 legislators

0.86

 congress

0.84

 Congress

0.84

 parliament

0.81

 politicians

0.79

 ministers

0.78

 governors

0.78

Activations Density 0.044%

phrases that indicate formal calls to action or requests for responses

No Comments

No Known Activations