INDEX

Explanations

mentions of distance or comparison in terms of extent or degree

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 fold

-0.68

 Benefits

-0.66

lez

-0.66

borough

-0.65

chio

-0.64

vation

-0.63

 Cosponsors

-0.62

 folds

-0.61

 EntityItem

-0.61

Micro

-0.61

POSITIVE LOGITS

 indications

0.74

 anyway

0.73

 hasn

0.72

 haven

0.70

 unanswered

0.69

 unexpl

0.69

 only

0.68

 provisional

0.67

 unsuccessful

0.66

 undisclosed

0.65

Activations Density 0.972%

mentions of distance or comparison in terms of extent or degree

No Comments

No Known Activations