developer_uid: Azazelle
submission_id: azazelle-l3-persephone-8_4975_v2
model_name: azazelle-l3-persephone-8_4975_v2
model_group: Azazelle/L3-Persephone-8
status: torndown
timestamp: 2024-07-05T18:21:46+00:00
num_battles: 32203
num_wins: 17122
celo_rating: 1200.67
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-8_4975_v2
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: 0.5316895941371922
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-v2-mkmlizer
Waiting for job on azazelle-l3-persephone-8-4975-v2-mkmlizer to finish
azazelle-l3-persephone-8-4975-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ _____ __ __ ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ /___/ ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ Version: 0.8.14 ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ https://mk1.ai ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ The license key for the current software has been verified as ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ belonging to: ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ Chai Research Corp. ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ║ ║
azazelle-l3-persephone-8-4975-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Retrying (%r) after connection broken by '%r': %s
azazelle-l3-persephone-8-4975-v2-mkmlizer: Downloaded to shared memory in 67.755s
azazelle-l3-persephone-8-4975-v2-mkmlizer: quantizing model to /dev/shm/model_cache
azazelle-l3-persephone-8-4975-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Retrying (%r) after connection broken by '%r': %s
azazelle-l3-persephone-8-4975-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<11:54, 2.47s/it] Loading 0: 2%|▏ | 6/291 [00:05<03:09, 1.50it/s] Loading 0: 3%|▎ | 10/291 [00:05<01:42, 2.73it/s] Loading 0: 5%|▌ | 15/291 [00:05<00:55, 4.93it/s] Loading 0: 8%|▊ | 23/291 [00:05<00:27, 9.62it/s] Loading 0: 11%|█ | 31/291 [00:05<00:16, 15.29it/s] Loading 0: 13%|█▎ | 37/291 [00:05<00:12, 19.78it/s] Loading 0: 15%|█▍ | 43/291 [00:05<00:10, 24.18it/s] Loading 0: 17%|█▋ | 50/291 [00:06<00:07, 30.69it/s] Loading 0: 20%|█▉ | 58/291 [00:06<00:06, 38.73it/s] Loading 0: 22%|██▏ | 65/291 [00:06<00:07, 29.57it/s] Loading 0: 24%|██▍ | 70/291 [00:06<00:06, 31.79it/s] Loading 0: 26%|██▋ | 77/291 [00:06<00:05, 37.92it/s] Loading 0: 29%|██▉ | 85/291 [00:06<00:04, 45.11it/s] Loading 0: 31%|███▏ | 91/291 [00:07<00:04, 47.56it/s] Loading 0: 33%|███▎ | 97/291 [00:07<00:04, 47.52it/s] Loading 0: 36%|███▌ | 104/291 [00:07<00:03, 51.81it/s] Loading 0: 38%|███▊ | 112/291 [00:07<00:03, 56.80it/s] Loading 0: 41%|████ | 119/291 [00:07<00:04, 35.74it/s] Loading 0: 43%|████▎ | 124/291 [00:07<00:04, 37.05it/s] Loading 0: 45%|████▌ | 131/291 [00:07<00:03, 42.89it/s] Loading 0: 48%|████▊ | 139/291 [00:08<00:03, 49.78it/s] Loading 0: 50%|████▉ | 145/291 [00:08<00:02, 51.18it/s] Loading 0: 52%|█████▏ | 151/291 [00:08<00:02, 50.17it/s] Loading 0: 54%|█████▍ | 158/291 [00:08<00:02, 54.57it/s] Loading 0: 57%|█████▋ | 166/291 [00:08<00:03, 36.85it/s] Loading 0: 60%|██████ | 175/291 [00:08<00:02, 44.23it/s] Loading 0: 62%|██████▏ | 181/291 [00:08<00:02, 47.23it/s] Loading 0: 64%|██████▍ | 187/291 [00:09<00:02, 47.62it/s] Loading 0: 67%|██████▋ | 194/291 [00:09<00:01, 52.25it/s] Loading 0: 69%|██████▉ | 202/291 [00:09<00:01, 57.66it/s] Loading 0: 72%|███████▏ | 209/291 [00:09<00:01, 60.39it/s] Loading 0: 74%|███████▍ | 216/291 [00:09<00:02, 36.84it/s] Loading 0: 76%|███████▋ | 222/291 [00:09<00:01, 38.27it/s] Loading 0: 79%|███████▉ | 230/291 [00:10<00:01, 45.24it/s] Loading 0: 82%|████████▏ | 238/291 [00:10<00:01, 50.78it/s] Loading 0: 84%|████████▍ | 244/291 [00:10<00:00, 51.38it/s] Loading 0: 86%|████████▌ | 250/291 [00:10<00:00, 48.89it/s] Loading 0: 88%|████████▊ | 257/291 [00:10<00:00, 52.24it/s] Loading 0: 91%|█████████ | 265/291 [00:10<00:00, 57.10it/s] Loading 0: 93%|█████████▎| 271/291 [00:11<00:00, 34.80it/s] Loading 0: 95%|█████████▍| 276/291 [00:11<00:00, 35.68it/s] Loading 0: 98%|█████████▊| 284/291 [00:11<00:00, 43.39it/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-v2-mkmlizer: quantized model in 33.619s
azazelle-l3-persephone-8-4975-v2-mkmlizer: Processed model Azazelle/L3-Persephone-8B-v1.0 in 101.374s
azazelle-l3-persephone-8-4975-v2-mkmlizer: creating bucket guanaco-mkml-models
azazelle-l3-persephone-8-4975-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
azazelle-l3-persephone-8-4975-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v2
azazelle-l3-persephone-8-4975-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v2/config.json
azazelle-l3-persephone-8-4975-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v2/special_tokens_map.json
azazelle-l3-persephone-8-4975-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v2/tokenizer_config.json
azazelle-l3-persephone-8-4975-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v2/tokenizer.json
azazelle-l3-persephone-8-4975-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v2/flywheel_model.0.safetensors
azazelle-l3-persephone-8-4975-v2-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
azazelle-l3-persephone-8-4975-v2-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-v2-mkmlizer: warnings.warn(
azazelle-l3-persephone-8-4975-v2-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-v2-mkmlizer: warnings.warn(
azazelle-l3-persephone-8-4975-v2-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-v2-mkmlizer: warnings.warn(
azazelle-l3-persephone-8-4975-v2-mkmlizer: /opt/conda/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
azazelle-l3-persephone-8-4975-v2-mkmlizer: return self.fget.__get__(instance, owner)()
azazelle-l3-persephone-8-4975-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
azazelle-l3-persephone-8-4975-v2-mkmlizer: Saving duration: 0.412s
azazelle-l3-persephone-8-4975-v2-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.377s
azazelle-l3-persephone-8-4975-v2-mkmlizer: creating bucket guanaco-reward-models
azazelle-l3-persephone-8-4975-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
azazelle-l3-persephone-8-4975-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v2_reward
azazelle-l3-persephone-8-4975-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v2_reward/config.json
azazelle-l3-persephone-8-4975-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v2_reward/tokenizer_config.json
azazelle-l3-persephone-8-4975-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v2_reward/special_tokens_map.json
azazelle-l3-persephone-8-4975-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v2_reward/merges.txt
azazelle-l3-persephone-8-4975-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v2_reward/vocab.json
azazelle-l3-persephone-8-4975-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v2_reward/tokenizer.json
azazelle-l3-persephone-8-4975-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v2_reward/reward.tensors
Job azazelle-l3-persephone-8-4975-v2-mkmlizer completed after 124.68s with status: succeeded
Stopping job with name azazelle-l3-persephone-8-4975-v2-mkmlizer
Pipeline stage MKMLizer completed in 125.61s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service azazelle-l3-persephone-8-4975-v2
Waiting for inference service azazelle-l3-persephone-8-4975-v2 to be ready
Inference service azazelle-l3-persephone-8-4975-v2 ready after 40.269418716430664s
Pipeline stage ISVCDeployer completed in 47.03s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.0928547382354736s
Received healthy response to inference request in 1.3867239952087402s
Received healthy response to inference request in 1.3259303569793701s
Received healthy response to inference request in 1.2738022804260254s
Received healthy response to inference request in 1.3422293663024902s
5 requests
0 failed requests
5th percentile: 1.2842278957366944
10th percentile: 1.2946535110473634
20th percentile: 1.3155047416687011
30th percentile: 1.329190158843994
40th percentile: 1.3357097625732421
50th percentile: 1.3422293663024902
60th percentile: 1.3600272178649901
70th percentile: 1.3778250694274903
80th percentile: 1.527950143814087
90th percentile: 1.8104024410247803
95th percentile: 1.9516285896301269
99th percentile: 2.0646095085144043
mean time: 1.4843081474304198
Pipeline stage StressChecker completed in 8.24s
azazelle-l3-persephone-8_4975_v2 status is now deployed due to DeploymentManager action
azazelle-l3-persephone-8_4975_v2 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of azazelle-l3-persephone-8_4975_v2
Running pipeline stage ISVCDeleter
Checking if service azazelle-l3-persephone-8-4975-v2 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 6.54s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key azazelle-l3-persephone-8-4975-v2/config.json from bucket guanaco-mkml-models
Deleting key azazelle-l3-persephone-8-4975-v2/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key azazelle-l3-persephone-8-4975-v2/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key azazelle-l3-persephone-8-4975-v2/tokenizer.json from bucket guanaco-mkml-models
Deleting key azazelle-l3-persephone-8-4975-v2/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key azazelle-l3-persephone-8-4975-v2_reward/config.json from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v2_reward/merges.txt from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v2_reward/reward.tensors from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v2_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v2_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v2_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v2_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.65s
azazelle-l3-persephone-8_4975_v2 status is now torndown due to DeploymentManager action