INDEX

Explanations

instances where someone or something is leaving a particular place or situation

oai_token-act-pair · gpt-3.5-turbo

instances of the word "leaving."

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

ought

-0.67

hatt

-0.63

als

-0.61

oust

-0.60

bled

-0.59

raphic

-0.58

hub

-0.58

fest

-0.57

NRS

-0.57

Ã»

-0.56

POSITIVE LOGITS

 leaving

3.20

 Leaving

2.07

 departing

1.94

 exiting

1.86

 leave

1.78

 abandoning

1.74

 quitting

1.65

leave

1.52

 entering

1.47

 leaves

1.45

Activations Density 0.020%

instances where someone or something is leaving a particular place or situation

instances of the word "leaving."

No Comments

No Known Activations