developer_uid: robert_irvine
submission_id: nousresearch-meta-llama_4941_v71
model_name: nousresearch-meta-llama_4941_v71
model_group: NousResearch/Meta-Llama-
status: torndown
timestamp: 2024-07-03T18:21:08+00:00
num_battles: 14378
num_wins: 6827
celo_rating: 1153.46
family_friendly_score: 0.0
submission_type: basic
model_repo: NousResearch/Meta-Llama-3-8B-Instruct
model_architecture: LlamaForCausalLM
reward_repo: rirv938/reward_gpt2_medium_preference_24m_e2
model_num_parameters: 8030261248.0
best_of: 4
max_input_tokens: 512
max_output_tokens: 64
display_name: nousresearch-meta-llama_4941_v71
is_internal_developer: True
language_model: NousResearch/Meta-Llama-3-8B-Instruct
model_size: 8B
ranking_group: single
us_pacific_date: 2024-07-03
win_ratio: 0.4748226457087217
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['</s>', '<|user|>', '###', '\n'], 'max_input_tokens': 512, 'best_of': 4, 'max_output_tokens': 64}
formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
reward_formatter: {'bot_template': 'Bot: {message}\n', 'memory_template': 'Memory: {memory}\n', 'prompt_template': '{prompt}\n', 'response_template': 'Bot:', 'truncate_by_message': False, 'user_template': 'User: {message}\n'}
Resubmit model
Running pipeline stage MKMLizer
Starting job with name nousresearch-meta-llama-4941-v71-mkmlizer
Waiting for job on nousresearch-meta-llama-4941-v71-mkmlizer to finish
nousresearch-meta-llama-4941-v71-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
nousresearch-meta-llama-4941-v71-mkmlizer: ║     _____            __           __                                ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║       /___/                                                         ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║                                                                     ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║  Version: 0.8.14                                                    ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║  https://mk1.ai                                                     ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║                                                                     ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║  The license key for the current software has been verified as      ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║  belonging to:                                                      ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║                                                                     ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║  Chai Research Corp.                                                ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
nousresearch-meta-llama-4941-v71-mkmlizer: ║                                                                     ║
nousresearch-meta-llama-4941-v71-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
nousresearch-meta-llama-4941-v71-mkmlizer: Downloaded to shared memory in 21.553s
nousresearch-meta-llama-4941-v71-mkmlizer: quantizing model to /dev/shm/model_cache
nousresearch-meta-llama-4941-v71-mkmlizer: Saving flywheel model at /dev/shm/model_cache
nousresearch-meta-llama-4941-v71-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   4%|▍         | 12/291 [00:00<00:02, 113.78it/s]
Loading 0:   8%|▊         | 24/291 [00:00<00:02, 101.54it/s]
Loading 0:  13%|█▎        | 37/291 [00:00<00:02, 112.61it/s]
Loading 0:  17%|█▋        | 49/291 [00:00<00:02, 104.91it/s]
Loading 0:  21%|██        | 60/291 [00:00<00:02, 104.84it/s]
Loading 0:  26%|██▌       | 75/291 [00:00<00:01, 112.92it/s]
Loading 0:  30%|██▉       | 87/291 [00:01<00:03, 59.77it/s] 
Loading 0:  35%|███▌      | 102/291 [00:01<00:02, 73.86it/s]
Loading 0:  38%|███▊      | 112/291 [00:01<00:02, 78.79it/s]
Loading 0:  42%|████▏     | 122/291 [00:01<00:02, 81.43it/s]
Loading 0:  46%|████▌     | 134/291 [00:01<00:01, 90.51it/s]
Loading 0:  51%|█████     | 147/291 [00:01<00:01, 98.68it/s]
Loading 0:  54%|█████▍    | 158/291 [00:01<00:01, 96.63it/s]
Loading 0:  59%|█████▉    | 171/291 [00:01<00:01, 104.37it/s]
Loading 0:  63%|██████▎   | 183/291 [00:01<00:01, 102.86it/s]
Loading 0:  67%|██████▋   | 194/291 [00:02<00:01, 59.74it/s] 
Loading 0:  70%|██████▉   | 203/291 [00:02<00:01, 64.71it/s]
Loading 0:  74%|███████▍  | 216/291 [00:02<00:00, 77.44it/s]
Loading 0:  79%|███████▊  | 229/291 [00:02<00:00, 85.86it/s]
Loading 0:  83%|████████▎ | 242/291 [00:02<00:00, 94.79it/s]
Loading 0:  88%|████████▊ | 255/291 [00:02<00:00, 98.34it/s]
Loading 0:  91%|█████████▏| 266/291 [00:03<00:00, 91.52it/s]
Loading 0:  95%|█████████▌| 277/291 [00:03<00:00, 95.94it/s]
Loading 0:  99%|█████████▉| 288/291 [00:08<00:00,  6.39it/s]
                                                            
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
nousresearch-meta-llama-4941-v71-mkmlizer: quantized model in 24.483s
nousresearch-meta-llama-4941-v71-mkmlizer: Processed model NousResearch/Meta-Llama-3-8B-Instruct in 46.036s
nousresearch-meta-llama-4941-v71-mkmlizer: creating bucket guanaco-mkml-models
nousresearch-meta-llama-4941-v71-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
nousresearch-meta-llama-4941-v71-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v71
nousresearch-meta-llama-4941-v71-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v71/config.json
nousresearch-meta-llama-4941-v71-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v71/tokenizer_config.json
nousresearch-meta-llama-4941-v71-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v71/special_tokens_map.json
nousresearch-meta-llama-4941-v71-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v71/tokenizer.json
nousresearch-meta-llama-4941-v71-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v71/flywheel_model.0.safetensors
nousresearch-meta-llama-4941-v71-mkmlizer: loading reward model from rirv938/reward_gpt2_medium_preference_24m_e2
nousresearch-meta-llama-4941-v71-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.
nousresearch-meta-llama-4941-v71-mkmlizer:   warnings.warn(
nousresearch-meta-llama-4941-v71-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`.
nousresearch-meta-llama-4941-v71-mkmlizer:   warnings.warn(
nousresearch-meta-llama-4941-v71-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.
nousresearch-meta-llama-4941-v71-mkmlizer:   warnings.warn(
nousresearch-meta-llama-4941-v71-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.
nousresearch-meta-llama-4941-v71-mkmlizer:   warnings.warn(
nousresearch-meta-llama-4941-v71-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()
nousresearch-meta-llama-4941-v71-mkmlizer:   return self.fget.__get__(instance, owner)()
nousresearch-meta-llama-4941-v71-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
nousresearch-meta-llama-4941-v71-mkmlizer: Saving duration: 0.443s
nousresearch-meta-llama-4941-v71-mkmlizer: Processed model rirv938/reward_gpt2_medium_preference_24m_e2 in 6.877s
nousresearch-meta-llama-4941-v71-mkmlizer: creating bucket guanaco-reward-models
nousresearch-meta-llama-4941-v71-mkmlizer: Bucket 's3://guanaco-reward-models/' created
nousresearch-meta-llama-4941-v71-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/nousresearch-meta-llama-4941-v71_reward
nousresearch-meta-llama-4941-v71-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v71_reward/config.json
nousresearch-meta-llama-4941-v71-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/nousresearch-meta-llama-4941-v71_reward/merges.txt
nousresearch-meta-llama-4941-v71-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v71_reward/tokenizer_config.json
nousresearch-meta-llama-4941-v71-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v71_reward/special_tokens_map.json
nousresearch-meta-llama-4941-v71-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v71_reward/tokenizer.json
nousresearch-meta-llama-4941-v71-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v71_reward/vocab.json
nousresearch-meta-llama-4941-v71-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/nousresearch-meta-llama-4941-v71_reward/reward.tensors
Job nousresearch-meta-llama-4941-v71-mkmlizer completed after 73.69s with status: succeeded
Stopping job with name nousresearch-meta-llama-4941-v71-mkmlizer
Pipeline stage MKMLizer completed in 74.66s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.12s
Running pipeline stage ISVCDeployer
Creating inference service nousresearch-meta-llama-4941-v71
Waiting for inference service nousresearch-meta-llama-4941-v71 to be ready
Inference service nousresearch-meta-llama-4941-v71 ready after 30.219446897506714s
Pipeline stage ISVCDeployer completed in 37.07s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.889639139175415s
Received healthy response to inference request in 1.1171624660491943s
Received healthy response to inference request in 0.7733283042907715s
Received healthy response to inference request in 1.1137490272521973s
Received healthy response to inference request in 1.1359047889709473s
5 requests
0 failed requests
5th percentile: 0.8414124488830567
10th percentile: 0.9094965934753418
20th percentile: 1.0456648826599122
30th percentile: 1.1144317150115968
40th percentile: 1.1157970905303956
50th percentile: 1.1171624660491943
60th percentile: 1.1246593952178956
70th percentile: 1.1321563243865966
80th percentile: 1.2866516590118409
90th percentile: 1.588145399093628
95th percentile: 1.7388922691345214
99th percentile: 1.8594897651672364
mean time: 1.2059567451477051
Pipeline stage StressChecker completed in 7.01s
nousresearch-meta-llama_4941_v71 status is now deployed due to DeploymentManager action
nousresearch-meta-llama_4941_v71 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of nousresearch-meta-llama_4941_v71
Running pipeline stage ISVCDeleter
Checking if service nousresearch-meta-llama-4941-v71 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 4.01s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key nousresearch-meta-llama-4941-v71/config.json from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v71/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v71/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v71/tokenizer.json from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v71/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key nousresearch-meta-llama-4941-v71_reward/config.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v71_reward/merges.txt from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v71_reward/reward.tensors from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v71_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v71_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v71_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v71_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.53s
nousresearch-meta-llama_4941_v71 status is now torndown due to DeploymentManager action