INDEX

Explanations

dates in the year

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

hed

-0.68

 ancest

-0.68

este

-0.68

ube

-0.66

ierrez

-0.66

bars

-0.65

phy

-0.64

ype

-0.63

hes

-0.62

abulary

-0.61

POSITIVE LOGITS

å¹

1.31

-'

1.06

 onwards

1.00

 edition

0.93

0.86

 onward

0.85

âĢ²

0.81

âĢ³

0.76

 election

0.75

 season

0.75

Activations Density 0.777%

dates in the year

No Comments

No Known Activations