INDEX

Explanations

references to organizations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

ILLE

-0.74

 Cheney

-0.71

PLIED

-0.71

ï¸ı

-0.66

nces

-0.66

 Pelicans

-0.66

avorite

-0.65

Aub

-0.62

 Christie

-0.62

Eag

-0.61

POSITIVE LOGITS

izational

1.73

isations

1.37

izations

1.31

izers

1.24

izer

1.15

ization

1.15

ized

1.14

isers

1.11

izing

1.08

izes

1.04

Activations Density 0.005%

references to organizations

No Comments

No Known Activations