INDEX

Explanations

references to beginners and introductory concepts 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 @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.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

 indefinite

-0.69

 arbitrary

-0.65

 veto

-0.65

rored

-0.63

 assass

-0.61

bleacher

-0.60

 likeness

-0.59

rid

-0.59

 retweet

-0.58

 Abbott

-0.58

POSITIVE LOGITS

 learners

0.85

Course

0.82

 Begin

0.82

 learning

0.81

 beginners

0.79

learning

0.79

 understanding

0.78

 beginner

0.78

 Learning

0.77

Learning

0.75

Activations Density 0.138%

references to beginners and introductory concepts in various contexts

No Comments

No Known Activations