INDEX

Explanations

phrases related to criminal activities or incidents

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

ancial

-0.76

geries

-0.75

raints

-0.70

ourgeois

-0.67

raft

-0.65

opl

-0.65

inctions

-0.64

agonal

-0.63

eta

-0.63

entanyl

-0.63

POSITIVE LOGITS

 deems

1.20

 deem

1.15

 deemed

0.97

 considers

0.95

 sorely

0.92

 dearly

0.91

 dubbed

0.89

 describes

0.85

 termed

0.83

 believe

0.83

Activations Density 0.809%

phrases related to criminal activities or incidents

No Comments

No Known Activations