INDEX

Explanations

phrases or structures indicating interactions and relationships

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

ï¸ı

-0.71

gerald

-0.71

hari

-0.70

occup

-0.65

-0.63

ggle

-0.63

coll

-0.63

Luc

-0.62

ovation

-0.62

igmatic

-0.61

POSITIVE LOGITS

erous

0.73

 behalf

0.72

 regards

0.70

ãĤ»

0.68

 impunity

0.68

 money

0.66

 regard

0.66

 Foster

0.60

illions

0.59

 Assets

0.59

Activations Density 0.125%

phrases or structures indicating interactions and relationships

No Comments

No Known Activations