INDEX

Explanations

keywords or key phrases

oai_token-act-pair · gpt-3.5-turbo

key phrases or terms that indicate important concepts or takeaways

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

 amused

-0.64

 bene

-0.63

 Whedon

-0.63

 forgiveness

-0.62

 shrug

-0.62

 Bett

-0.61

 Vulcan

-0.61

 forgiving

-0.61

civ

-0.61

 aunt

-0.60

POSITIVE LOGITS

Key

3.81

key

2.48

KEY

2.46

Keys

2.37

Key

2.06

keys

1.85

KEY

1.79

key

1.73

 Keys

1.55

 keys

1.55

Activations Density 0.011%

keywords or key phrases

key phrases or terms that indicate important concepts or takeaways

No Comments

No Known Activations