INDEX

Explanations

phrases indicating placement or positioning of objects or people

oai_token-act-pair · gpt-3.5-turbo

instances of the word "placed."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

maxwell

-0.73

rick

-0.65

bart

-0.65

docs

-0.63

stay

-0.63

 filibuster

-0.61

lin

-0.61

ibus

-0.61

din

-0.59

 Wikipedia

-0.59

POSITIVE LOGITS

 placed

3.42

placed

1.97

 positioned

1.87

 placing

1.72

 inserted

1.63

 situated

1.47

 planted

1.44

 placement

1.44

 erected

1.41

 implanted

1.35

Activations Density 0.024%

phrases indicating placement or positioning of objects or people

instances of the word "placed."

No Comments

No Known Activations