INDEX

Explanations

phrases related to war conflicts, military actions, and terrorist 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_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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

payer

-0.79

 concess

-0.76

NAD

-0.76

NH

-0.75

 Pension

-0.72

 Canal

-0.72

annel

-0.71

 nutrit

-0.71

 Quality

-0.70

 Improvement

-0.69

POSITIVE LOGITS

 infiltr

1.33

 invading

1.24

 attacks

1.24

 attack

1.22

 infiltration

1.21

 terrorists

1.19

 menacing

1.18

 attacking

1.17

 kidn

1.16

 attackers

1.16

Activations Density 4.271%

phrases related to war conflicts, military actions, and terrorist groups

No Comments

No Known Activations