INDEX

Explanations

phrases or sentences that start with the word "Looking"

oai_token-act-pair · gpt-3.5-turbo

the phrase "Looking" indicating inquiries or searches for information and trustworthiness

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

 charge

-0.69

 penalties

-0.66

 override

-0.66

 penalty

-0.66

 invented

-0.65

 elect

-0.64

 chore

-0.63

 viol

-0.63

 punishment

-0.63

 itself

-0.62

POSITIVE LOGITS

Looking

3.32

 Looking

2.85

looking

1.89

Going

1.61

 looking

1.50

Seeing

1.45

 Going

1.36

 LOOK

1.32

 Thinking

1.31

look

1.28

Activations Density 0.014%

phrases or sentences that start with the word "Looking"

the phrase "Looking" indicating inquiries or searches for information and trustworthiness

No Comments

No Known Activations