INDEX

Explanations

information related to a news story or event

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

ndra

-0.87

aternity

-0.81

 alliances

-0.73

ablished

-0.72

anship

-0.70

 committees

-0.69

 attrition

-0.69

 hosp

-0.66

alties

-0.66

 priorities

-0.66

POSITIVE LOGITS

 footage

1.54

 uploaded

1.48

 video

1.45

 videos

1.44

 clips

1.33

 recordings

1.32

 Videos

1.31

 screenshots

1.30

 images

1.28

 VIDEO

1.26

Activations Density 2.330%

information related to a news story or event

No Comments

No Known Activations