developer_uid: Azazelle
submission_id: azazelle-l3-persephone-8_4975_v3
model_name: azazelle-l3-persephone-8b
model_group: Azazelle/L3-Persephone-8
status: torndown
timestamp: 2024-07-06T00:15:03+00:00
num_battles: 34066
num_wins: 17939
celo_rating: 1201.38
family_friendly_score: 0.0
submission_type: basic
model_repo: Azazelle/L3-Persephone-8B-v1.0
model_architecture: LlamaForCausalLM
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: azazelle-l3-persephone-8b
is_internal_developer: False
language_model: Azazelle/L3-Persephone-8B-v1.0
model_size: 8B
ranking_group: single
us_pacific_date: 2024-07-05
win_ratio: 0.5265954323959373
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.08, 'top_k': 50, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<eot_id>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}
formatter: {'memory_template': "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\n\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}:', 'truncate_by_message': False}
reward_formatter: {'bot_template': '{bot_name}: {message}\n', 'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'response_template': '{bot_name}:', 'truncate_by_message': False, 'user_template': '{user_name}: {message}\n'}
Resubmit model
Running pipeline stage MKMLizer
Starting job with name azazelle-l3-persephone-8-4975-v3-mkmlizer
Waiting for job on azazelle-l3-persephone-8-4975-v3-mkmlizer to finish
azazelle-l3-persephone-8-4975-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ _____ __ __ ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ /___/ ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ Version: 0.8.14 ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ https://mk1.ai ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ The license key for the current software has been verified as ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ belonging to: ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ Chai Research Corp. ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ║ ║
azazelle-l3-persephone-8-4975-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
azazelle-l3-persephone-8-4975-v3-mkmlizer: Downloaded to shared memory in 52.255s
azazelle-l3-persephone-8-4975-v3-mkmlizer: quantizing model to /dev/shm/model_cache
azazelle-l3-persephone-8-4975-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
azazelle-l3-persephone-8-4975-v3-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:05<12:29, 2.59s/it] Loading 0: 2%|▏ | 6/291 [00:05<03:18, 1.44it/s] Loading 0: 3%|▎ | 10/291 [00:05<01:46, 2.64it/s] Loading 0: 5%|▌ | 15/291 [00:05<00:57, 4.79it/s] Loading 0: 8%|▊ | 23/291 [00:05<00:28, 9.36it/s] Loading 0: 11%|█ | 31/291 [00:05<00:17, 14.90it/s] Loading 0: 13%|█▎ | 37/291 [00:06<00:13, 19.31it/s] Loading 0: 15%|█▍ | 43/291 [00:06<00:10, 23.57it/s] Loading 0: 17%|█▋ | 50/291 [00:06<00:08, 29.94it/s] Loading 0: 20%|█▉ | 58/291 [00:06<00:06, 37.83it/s] Loading 0: 22%|██▏ | 65/291 [00:06<00:08, 28.23it/s] Loading 0: 24%|██▍ | 70/291 [00:06<00:07, 30.42it/s] Loading 0: 26%|██▋ | 77/291 [00:07<00:05, 36.68it/s] Loading 0: 29%|██▉ | 85/291 [00:07<00:04, 44.16it/s] Loading 0: 31%|███▏ | 91/291 [00:07<00:04, 45.75it/s] Loading 0: 33%|███▎ | 97/291 [00:07<00:04, 45.51it/s] Loading 0: 36%|███▌ | 104/291 [00:07<00:03, 49.36it/s] Loading 0: 38%|███▊ | 112/291 [00:07<00:03, 54.56it/s] Loading 0: 41%|████ | 118/291 [00:08<00:05, 31.74it/s] Loading 0: 42%|████▏ | 123/291 [00:08<00:05, 32.99it/s] Loading 0: 45%|████▌ | 131/291 [00:08<00:03, 40.28it/s] Loading 0: 48%|████▊ | 139/291 [00:08<00:03, 46.69it/s] Loading 0: 50%|████▉ | 145/291 [00:08<00:03, 47.95it/s] Loading 0: 52%|█████▏ | 151/291 [00:08<00:02, 47.19it/s] Loading 0: 54%|█████▍ | 158/291 [00:08<00:02, 50.62it/s] Loading 0: 57%|█████▋ | 166/291 [00:09<00:03, 34.39it/s] Loading 0: 59%|█████▉ | 172/291 [00:09<00:03, 38.73it/s] Loading 0: 61%|██████ | 177/291 [00:09<00:02, 40.19it/s] Loading 0: 64%|██████▍ | 186/291 [00:09<00:02, 47.73it/s] Loading 0: 67%|██████▋ | 195/291 [00:09<00:01, 53.51it/s] Loading 0: 70%|██████▉ | 203/291 [00:09<00:01, 59.59it/s] Loading 0: 73%|███████▎ | 211/291 [00:09<00:01, 64.43it/s] Loading 0: 75%|███████▍ | 218/291 [00:10<00:01, 39.68it/s] Loading 0: 77%|███████▋ | 224/291 [00:10<00:01, 41.62it/s] Loading 0: 79%|███████▉ | 230/291 [00:10<00:01, 44.53it/s] Loading 0: 82%|████████▏ | 238/291 [00:10<00:01, 50.58it/s] Loading 0: 84%|████████▍ | 244/291 [00:10<00:00, 51.16it/s] Loading 0: 86%|████████▌ | 250/291 [00:10<00:00, 49.12it/s] Loading 0: 88%|████████▊ | 257/291 [00:10<00:00, 52.04it/s] Loading 0: 91%|█████████ | 265/291 [00:11<00:00, 57.01it/s] Loading 0: 93%|█████████▎| 271/291 [00:11<00:00, 33.66it/s] Loading 0: 95%|█████████▍| 276/291 [00:11<00:00, 34.24it/s] Loading 0: 98%|█████████▊| 284/291 [00:11<00:00, 41.66it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
azazelle-l3-persephone-8-4975-v3-mkmlizer: quantized model in 34.058s
azazelle-l3-persephone-8-4975-v3-mkmlizer: Processed model Azazelle/L3-Persephone-8B-v1.0 in 86.314s
azazelle-l3-persephone-8-4975-v3-mkmlizer: creating bucket guanaco-mkml-models
azazelle-l3-persephone-8-4975-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
azazelle-l3-persephone-8-4975-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v3
azazelle-l3-persephone-8-4975-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v3/special_tokens_map.json
azazelle-l3-persephone-8-4975-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v3/config.json
azazelle-l3-persephone-8-4975-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v3/tokenizer_config.json
azazelle-l3-persephone-8-4975-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v3/tokenizer.json
azazelle-l3-persephone-8-4975-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v3/flywheel_model.0.safetensors
azazelle-l3-persephone-8-4975-v3-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
azazelle-l3-persephone-8-4975-v3-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:919: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
azazelle-l3-persephone-8-4975-v3-mkmlizer: warnings.warn(
azazelle-l3-persephone-8-4975-v3-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
azazelle-l3-persephone-8-4975-v3-mkmlizer: warnings.warn(
azazelle-l3-persephone-8-4975-v3-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:769: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
azazelle-l3-persephone-8-4975-v3-mkmlizer: warnings.warn(
azazelle-l3-persephone-8-4975-v3-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:468: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
azazelle-l3-persephone-8-4975-v3-mkmlizer: warnings.warn(
Connection pool is full, discarding connection: %s
Job azazelle-l3-persephone-8-4975-v3-mkmlizer completed after 155.21s with status: succeeded
Stopping job with name azazelle-l3-persephone-8-4975-v3-mkmlizer
Pipeline stage MKMLizer completed in 156.09s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service azazelle-l3-persephone-8-4975-v3
Waiting for inference service azazelle-l3-persephone-8-4975-v3 to be ready
Inference service azazelle-l3-persephone-8-4975-v3 ready after 40.18581771850586s
Pipeline stage ISVCDeployer completed in 46.83s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.0508170127868652s
Received healthy response to inference request in 1.4624712467193604s
Received healthy response to inference request in 1.3220744132995605s
Received healthy response to inference request in 1.2942473888397217s
Received healthy response to inference request in 1.3265364170074463s
5 requests
0 failed requests
5th percentile: 1.2998127937316895
10th percentile: 1.3053781986236572
20th percentile: 1.3165090084075928
30th percentile: 1.3229668140411377
40th percentile: 1.324751615524292
50th percentile: 1.3265364170074463
60th percentile: 1.3809103488922119
70th percentile: 1.4352842807769775
80th percentile: 1.5801403999328614
90th percentile: 1.8154787063598634
95th percentile: 1.9331478595733642
99th percentile: 2.027283182144165
mean time: 1.4912292957305908
Pipeline stage StressChecker completed in 8.09s
azazelle-l3-persephone-8_4975_v3 status is now deployed due to DeploymentManager action
azazelle-l3-persephone-8_4975_v3 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of azazelle-l3-persephone-8_4975_v3
Running pipeline stage ISVCDeleter
Checking if service azazelle-l3-persephone-8-4975-v3 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 6.02s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key azazelle-l3-persephone-8-4975-v3/config.json from bucket guanaco-mkml-models
Deleting key azazelle-l3-persephone-8-4975-v3/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key azazelle-l3-persephone-8-4975-v3/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key azazelle-l3-persephone-8-4975-v3/tokenizer.json from bucket guanaco-mkml-models
Deleting key azazelle-l3-persephone-8-4975-v3/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key azazelle-l3-persephone-8-4975-v3_reward/config.json from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v3_reward/merges.txt from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v3_reward/reward.tensors from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v3_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v3_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v3_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v3_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.26s
azazelle-l3-persephone-8_4975_v3 status is now torndown due to DeploymentManager action