developer_uid: Azazelle
submission_id: azazelle-l3-persephone-8_4975_v4
model_name: azazelle-l3-persephone-8b
model_group: Azazelle/L3-Persephone-8
status: torndown
timestamp: 2024-07-06T00:15:36+00:00
num_battles: 34225
num_wins: 17867
celo_rating: 1198.11
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: 0.522045288531775
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-v4-mkmlizer
Waiting for job on azazelle-l3-persephone-8-4975-v4-mkmlizer to finish
azazelle-l3-persephone-8-4975-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ _____ __ __ ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ /___/ ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ Version: 0.8.14 ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ https://mk1.ai ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ The license key for the current software has been verified as ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ belonging to: ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ Chai Research Corp. ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ║ ║
azazelle-l3-persephone-8-4975-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
azazelle-l3-persephone-8-4975-v4-mkmlizer: Downloaded to shared memory in 25.580s
azazelle-l3-persephone-8-4975-v4-mkmlizer: quantizing model to /dev/shm/model_cache
azazelle-l3-persephone-8-4975-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
azazelle-l3-persephone-8-4975-v4-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<11:06, 2.31s/it] Loading 0: 2%|▏ | 6/291 [00:04<02:55, 1.62it/s] Loading 0: 3%|▎ | 10/291 [00:04<01:30, 3.09it/s] Loading 0: 5%|▌ | 15/291 [00:05<00:49, 5.59it/s] Loading 0: 8%|▊ | 24/291 [00:05<00:23, 11.36it/s] Loading 0: 11%|█▏ | 33/291 [00:05<00:14, 18.09it/s] Loading 0: 14%|█▍ | 42/291 [00:05<00:09, 25.34it/s] Loading 0: 18%|█▊ | 51/291 [00:05<00:07, 32.87it/s] Loading 0: 21%|██ | 60/291 [00:05<00:06, 33.36it/s] Loading 0: 24%|██▎ | 69/291 [00:05<00:05, 40.21it/s] Loading 0: 27%|██▋ | 78/291 [00:06<00:04, 46.69it/s] Loading 0: 30%|██▉ | 87/291 [00:06<00:03, 51.82it/s] Loading 0: 33%|███▎ | 96/291 [00:06<00:03, 56.28it/s] Loading 0: 36%|███▌ | 105/291 [00:06<00:03, 59.69it/s] Loading 0: 39%|███▉ | 113/291 [00:06<00:03, 49.92it/s] Loading 0: 42%|████▏ | 122/291 [00:06<00:03, 54.76it/s] Loading 0: 45%|████▌ | 131/291 [00:06<00:02, 58.09it/s] Loading 0: 48%|████▊ | 140/291 [00:07<00:02, 61.14it/s] Loading 0: 51%|█████ | 149/291 [00:07<00:02, 63.39it/s] Loading 0: 54%|█████▍ | 158/291 [00:07<00:02, 63.86it/s] Loading 0: 57%|█████▋ | 166/291 [00:07<00:02, 51.63it/s] Loading 0: 59%|█████▉ | 173/291 [00:07<00:02, 55.30it/s] Loading 0: 62%|██████▏ | 180/291 [00:07<00:01, 56.19it/s] Loading 0: 64%|██████▍ | 186/291 [00:07<00:01, 54.48it/s] Loading 0: 67%|██████▋ | 195/291 [00:08<00:01, 58.63it/s] Loading 0: 70%|███████ | 204/291 [00:08<00:01, 61.92it/s] Loading 0: 73%|███████▎ | 213/291 [00:08<00:01, 49.55it/s] Loading 0: 76%|███████▋ | 222/291 [00:08<00:01, 54.40it/s] Loading 0: 79%|███████▉ | 231/291 [00:08<00:01, 58.07it/s] Loading 0: 82%|████████▏ | 240/291 [00:08<00:00, 61.07it/s] Loading 0: 86%|████████▌ | 249/291 [00:08<00:00, 63.68it/s] Loading 0: 89%|████████▊ | 258/291 [00:09<00:00, 65.57it/s] Loading 0: 91%|█████████▏| 266/291 [00:09<00:00, 51.92it/s] Loading 0: 95%|█████████▍| 275/291 [00:09<00:00, 58.17it/s] Loading 0: 98%|█████████▊| 284/291 [00:09<00:00, 64.02it/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-v4-mkmlizer: quantized model in 21.997s
azazelle-l3-persephone-8-4975-v4-mkmlizer: Processed model Azazelle/L3-Persephone-8B-v1.0 in 47.577s
azazelle-l3-persephone-8-4975-v4-mkmlizer: creating bucket guanaco-mkml-models
azazelle-l3-persephone-8-4975-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
azazelle-l3-persephone-8-4975-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v4
azazelle-l3-persephone-8-4975-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v4/special_tokens_map.json
azazelle-l3-persephone-8-4975-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v4/config.json
azazelle-l3-persephone-8-4975-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v4/tokenizer_config.json
azazelle-l3-persephone-8-4975-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v4/tokenizer.json
azazelle-l3-persephone-8-4975-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/azazelle-l3-persephone-8-4975-v4/flywheel_model.0.safetensors
azazelle-l3-persephone-8-4975-v4-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
azazelle-l3-persephone-8-4975-v4-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-v4-mkmlizer: warnings.warn(
azazelle-l3-persephone-8-4975-v4-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-v4-mkmlizer: warnings.warn(
azazelle-l3-persephone-8-4975-v4-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-v4-mkmlizer: warnings.warn(
azazelle-l3-persephone-8-4975-v4-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-v4-mkmlizer: warnings.warn(
azazelle-l3-persephone-8-4975-v4-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-v4-mkmlizer: return self.fget.__get__(instance, owner)()
azazelle-l3-persephone-8-4975-v4-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
azazelle-l3-persephone-8-4975-v4-mkmlizer: Saving duration: 0.240s
azazelle-l3-persephone-8-4975-v4-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 7.728s
azazelle-l3-persephone-8-4975-v4-mkmlizer: creating bucket guanaco-reward-models
azazelle-l3-persephone-8-4975-v4-mkmlizer: Bucket 's3://guanaco-reward-models/' created
azazelle-l3-persephone-8-4975-v4-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v4_reward
azazelle-l3-persephone-8-4975-v4-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v4_reward/config.json
azazelle-l3-persephone-8-4975-v4-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v4_reward/tokenizer_config.json
azazelle-l3-persephone-8-4975-v4-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v4_reward/special_tokens_map.json
azazelle-l3-persephone-8-4975-v4-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v4_reward/merges.txt
azazelle-l3-persephone-8-4975-v4-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v4_reward/vocab.json
azazelle-l3-persephone-8-4975-v4-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v4_reward/tokenizer.json
azazelle-l3-persephone-8-4975-v4-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/azazelle-l3-persephone-8-4975-v4_reward/reward.tensors
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Job azazelle-l3-persephone-8-4975-v4-mkmlizer completed after 142.96s with status: succeeded
Stopping job with name azazelle-l3-persephone-8-4975-v4-mkmlizer
Pipeline stage MKMLizer completed in 143.84s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.14s
Running pipeline stage ISVCDeployer
Creating inference service azazelle-l3-persephone-8-4975-v4
Waiting for inference service azazelle-l3-persephone-8-4975-v4 to be ready
Inference service azazelle-l3-persephone-8-4975-v4 ready after 50.258517265319824s
Pipeline stage ISVCDeployer completed in 57.73s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.107689142227173s
Received healthy response to inference request in 1.325019359588623s
Received healthy response to inference request in 1.474735975265503s
Received healthy response to inference request in 1.273557186126709s
Received healthy response to inference request in 1.338172435760498s
5 requests
0 failed requests
5th percentile: 1.2838496208190917
10th percentile: 1.2941420555114747
20th percentile: 1.3147269248962403
30th percentile: 1.327649974822998
40th percentile: 1.332911205291748
50th percentile: 1.338172435760498
60th percentile: 1.3927978515625
70th percentile: 1.447423267364502
80th percentile: 1.601326608657837
90th percentile: 1.854507875442505
95th percentile: 1.9810985088348387
99th percentile: 2.082371015548706
mean time: 1.5038348197937013
Pipeline stage StressChecker completed in 8.27s
azazelle-l3-persephone-8_4975_v4 status is now deployed due to DeploymentManager action
azazelle-l3-persephone-8_4975_v4 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of azazelle-l3-persephone-8_4975_v4
Running pipeline stage ISVCDeleter
Checking if service azazelle-l3-persephone-8-4975-v4 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 6.26s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key azazelle-l3-persephone-8-4975-v4/config.json from bucket guanaco-mkml-models
Deleting key azazelle-l3-persephone-8-4975-v4/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key azazelle-l3-persephone-8-4975-v4/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key azazelle-l3-persephone-8-4975-v4/tokenizer.json from bucket guanaco-mkml-models
Deleting key azazelle-l3-persephone-8-4975-v4/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key azazelle-l3-persephone-8-4975-v4_reward/config.json from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v4_reward/merges.txt from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v4_reward/reward.tensors from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v4_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v4_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v4_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key azazelle-l3-persephone-8-4975-v4_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.28s
azazelle-l3-persephone-8_4975_v4 status is now torndown due to DeploymentManager action