INDEX

Explanations

tennis players' names

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to notable individuals, particularly in sports or entertainment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.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

 disparate

-0.66

result

-0.60

 entitle

-0.58

 Seym

-0.57

 lesser

-0.56

CBO

-0.55

 magnification

-0.54

 Borders

-0.53

 Avalon

-0.53

bound

-0.53

POSITIVE LOGITS

Jr

0.91

igham

0.84

's

0.81

enegger

0.80

Sr

0.77

ovich

0.77

vic

0.76

eln

0.76

cci

0.74

 confid

0.71

Activations Density 0.192%

tennis players' names

proper nouns related to notable individuals, particularly in sports or entertainment

No Comments

No Known Activations