INDEX

Explanations

personal pronouns and possessive pronouns

oai_token-act-pair · gpt-3.5-turbo 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

 Mastery

-0.71

 Closed

-0.63

urated

-0.63

 Reaction

-0.63

EStreamFrame

-0.60

 Participation

-0.60

 Sched

-0.60

 Cutter

-0.59

 boards

-0.58

 Trident

-0.57

POSITIVE LOGITS

 harm

0.84

 justice

0.84

 injustice

0.73

 deserve

0.72

Ã±a

0.69

ught

0.67

tip

0.66

 trick

0.65

berra

0.64

 darn

0.64

Activations Density 0.127%

personal pronouns and possessive pronouns

No Comments

No Known Activations