INDEX

Explanations

negative emotions and uncertainties expressed through informal language and abbreviations

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

attery

-0.83

erity

-0.79

 sudden

-0.66

 prof

-0.64

ghazi

-0.64

cffff

-0.62

once

-0.59

pmwiki

-0.59

 coincided

-0.59

akespeare

-0.58

POSITIVE LOGITS

pees

0.79

 unanswered

0.70

TBA

0.68

 adolesc

0.67

 figuring

0.65

 infancy

0.65

 unresolved

0.65

UI

0.64

aints

0.63

ipop

0.63

Activations Density 0.186%

negative emotions and uncertainties expressed through informal language and abbreviations

No Comments

No Known Activations