INDEX

Explanations

mentions relating to public parks

oai_token-act-pair · gpt-3.5-turbo

references to parks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Brill

-0.68

tm

-0.66

 Tablet

-0.66

Î½

-0.65

htar

-0.64

 Volume

-0.61

imen

-0.61

 Brah

-0.61

 Disciple

-0.60

alm

-0.60

POSITIVE LOGITS

 parks

3.96

 park

2.57

 Parks

2.42

park

1.96

 Park

1.80

 PARK

1.65

 museums

1.61

 gardens

1.56

Park

1.55

 malls

1.52

Activations Density 0.009%

mentions relating to public parks

references to parks

No Comments

No Known Activations