INDEX

Explanations

words related to negative or harmful actions towards individuals or groups

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

istrates

-0.74

çīĪ

-0.73

 GOODMAN

-0.67

 giveaway

-0.67

ä»£

-0.66

HCR

-0.65

spot

-0.63

nesday

-0.62

åĮ

-0.61

mits

-0.61

POSITIVE LOGITS

aging

1.87

agement

1.76

aged

1.69

ages

1.61

age

1.49

AGE

1.47

AGES

1.27

agers

1.21

agements

1.15

ager

1.13

Activations Density 0.063%

words related to negative or harmful actions towards individuals or groups

No Comments

No Known Activations