INDEX

Explanations

examples of various concepts or items listed in a document

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "examples" in various contexts

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

deen

-0.71

 Strait

-0.70

 Bris

-0.59

ird

-0.59

 Lodge

-0.59

ught

-0.58

 bothering

-0.58

ospital

-0.57

erate

-0.57

ye

-0.56

POSITIVE LOGITS

 examples

3.72

 Examples

2.58

Examples

2.38

 example

2.37

 Example

1.93

example

1.83

 instances

1.80

Example

1.79

 exceptions

1.54

 illustrations

1.47

Activations Density 0.016%

examples of various concepts or items listed in a document

occurrences of the word "examples" in various contexts

No Comments

No Known Activations