INDEX

Explanations

phrases related to technology and digital content

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SPONSORED

-0.70

 otherwise

-0.66

present

-0.65

––

-0.63

such

-0.61

—"

-0.60

udeb

-0.59

than

-0.59

olson

-0.58

meaning

-0.57

POSITIVE LOGITS

 Matters

0.89

 Advice

0.86

 Problems

0.79

 Breaker

0.79

Girl

0.78

 Updates

0.78

Friend

0.76

 Selection

0.76

 Room

0.76

 Friendly

0.75

Activations Density 0.462%

phrases related to technology and digital content

No Comments

No Known Activations