INDEX

Explanations

references to specific years and economic trends

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

 hypocr

-0.74

avorite

-0.72

amina

-0.64

 astroph

-0.63

 resemb

-0.62

malink

-0.61

icio

-0.59

 VIDEOS

-0.58

 Sail

-0.58

achus

-0.56

POSITIVE LOGITS

 onwards

1.13

 onward

0.91

-'

0.90

oun

0.79

 inception

0.78

ãģ®éŃĶ

0.73

å¹

0.71

ãĥ©ãĥ³

0.67

ussen

0.65

âĢ²

0.65

Activations Density 0.038%

references to specific years and economic trends

No Comments

No Known Activations