INDEX

Explanations

instances of the word "approach"

oai_token-act-pair · gpt-3.5-turbo

references to various methodologies or strategies

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

netflix

-0.76

holes

-0.70

bucks

-0.70

ursed

-0.69

pelled

-0.64

 Advertisement

-0.64

Featured

-0.63

ogg

-0.63

 Gutenberg

-0.62

 Bones

-0.62

POSITIVE LOGITS

 approach

3.62

 Approach

2.73

 approaches

2.39

 attitude

1.67

 method

1.45

 methodology

1.40

 strategy

1.39

 approached

1.38

 approaching

1.36

 technique

1.36

Activations Density 0.022%

instances of the word "approach"

references to various methodologies or strategies

No Comments

No Known Activations