submission_id: nousresearch-meta-llama_4941_v69
developer_uid: robert_irvine
best_of: 1
celo_rating: 1113.6
display_name: nousresearch-meta-llama_4941_v69
family_friendly_score: 0.0
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}
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': 1, 'max_output_tokens': 64}
is_internal_developer: True
language_model: NousResearch/Meta-Llama-3-8B-Instruct
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: NousResearch/Meta-Llama-
model_name: nousresearch-meta-llama_4941_v69
model_num_parameters: 8030261248.0
model_repo: NousResearch/Meta-Llama-3-8B-Instruct
model_size: 8B
num_battles: 14313
num_wins: 5964
ranking_group: single
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'}
reward_repo: rirv938/reward_gpt2_medium_preference_24m_e2
status: torndown
submission_type: basic
timestamp: 2024-07-03T18:19:05+00:00
us_pacific_date: 2024-07-03
win_ratio: 0.41668413330538673
Resubmit model
Running pipeline stage MKMLizer
Starting job with name nousresearch-meta-llama-4941-v69-mkmlizer
Waiting for job on nousresearch-meta-llama-4941-v69-mkmlizer to finish
nousresearch-meta-llama-4941-v69-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
nousresearch-meta-llama-4941-v69-mkmlizer: ║     _____            __           __                                ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║       /___/                                                         ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║                                                                     ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║  Version: 0.8.14                                                    ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║  https://mk1.ai                                                     ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║                                                                     ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║  The license key for the current software has been verified as      ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║  belonging to:                                                      ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║                                                                     ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║  Chai Research Corp.                                                ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
nousresearch-meta-llama-4941-v69-mkmlizer: ║                                                                     ║
nousresearch-meta-llama-4941-v69-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
nousresearch-meta-llama-4941-v69-mkmlizer: Downloaded to shared memory in 21.930s
nousresearch-meta-llama-4941-v69-mkmlizer: quantizing model to /dev/shm/model_cache
nousresearch-meta-llama-4941-v69-mkmlizer: Saving flywheel model at /dev/shm/model_cache
nousresearch-meta-llama-4941-v69-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   1%|          | 3/291 [00:02<04:20,  1.10it/s]
Loading 0:   1%|▏         | 4/291 [00:06<08:04,  1.69s/it]
Loading 0:   2%|▏         | 5/291 [00:06<05:53,  1.24s/it]
Loading 0:   3%|▎         | 9/291 [00:06<02:11,  2.15it/s]
Loading 0:   4%|▍         | 12/291 [00:06<01:25,  3.25it/s]
Loading 0:   5%|▍         | 14/291 [00:07<01:14,  3.70it/s]
Loading 0:   7%|▋         | 21/291 [00:07<00:35,  7.66it/s]
Loading 0:   8%|▊         | 23/291 [00:07<00:32,  8.31it/s]
Loading 0:  13%|█▎        | 37/291 [00:07<00:11, 21.80it/s]
Loading 0:  17%|█▋        | 49/291 [00:07<00:07, 33.62it/s]
Loading 0:  21%|██▏       | 62/291 [00:07<00:04, 48.07it/s]
Loading 0:  26%|██▌       | 76/291 [00:07<00:03, 62.43it/s]
Loading 0:  30%|██▉       | 86/291 [00:08<00:04, 46.18it/s]
Loading 0:  35%|███▌      | 102/291 [00:08<00:03, 62.78it/s]
Loading 0:  39%|███▉      | 113/291 [00:08<00:02, 70.05it/s]
Loading 0:  44%|████▍     | 129/291 [00:08<00:01, 87.21it/s]
Loading 0:  48%|████▊     | 141/291 [00:08<00:01, 91.59it/s]
Loading 0:  54%|█████▎    | 156/291 [00:08<00:01, 103.56it/s]
Loading 0:  58%|█████▊    | 168/291 [00:08<00:01, 104.73it/s]
Loading 0:  62%|██████▏   | 181/291 [00:08<00:01, 108.87it/s]
Loading 0:  66%|██████▋   | 193/291 [00:09<00:01, 67.31it/s] 
Loading 0:  71%|███████   | 206/291 [00:09<00:01, 78.72it/s]
Loading 0:  75%|███████▌  | 219/291 [00:09<00:00, 89.04it/s]
Loading 0:  79%|███████▉  | 231/291 [00:09<00:00, 91.68it/s]
Loading 0:  85%|████████▍ | 246/291 [00:09<00:00, 102.12it/s]
Loading 0:  89%|████████▊ | 258/291 [00:09<00:00, 101.20it/s]
Loading 0:  94%|█████████▍| 273/291 [00:09<00:00, 109.60it/s]
Loading 0:  98%|█████████▊| 285/291 [00:10<00:00, 106.52it/s]
                                                             
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
nousresearch-meta-llama-4941-v69-mkmlizer: quantized model in 38.011s
nousresearch-meta-llama-4941-v69-mkmlizer: Processed model NousResearch/Meta-Llama-3-8B-Instruct in 59.942s
nousresearch-meta-llama-4941-v69-mkmlizer: creating bucket guanaco-mkml-models
nousresearch-meta-llama-4941-v69-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
nousresearch-meta-llama-4941-v69-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v69
nousresearch-meta-llama-4941-v69-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v69/config.json
nousresearch-meta-llama-4941-v69-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v69/special_tokens_map.json
nousresearch-meta-llama-4941-v69-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v69/tokenizer_config.json
nousresearch-meta-llama-4941-v69-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v69/tokenizer.json
nousresearch-meta-llama-4941-v69-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v69/flywheel_model.0.safetensors
nousresearch-meta-llama-4941-v69-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-v69-mkmlizer:   warnings.warn(
nousresearch-meta-llama-4941-v69-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-v69-mkmlizer:   warnings.warn(
nousresearch-meta-llama-4941-v69-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-v69-mkmlizer:   return self.fget.__get__(instance, owner)()
nousresearch-meta-llama-4941-v69-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
nousresearch-meta-llama-4941-v69-mkmlizer: Saving duration: 0.406s
nousresearch-meta-llama-4941-v69-mkmlizer: Processed model rirv938/reward_gpt2_medium_preference_24m_e2 in 7.668s
nousresearch-meta-llama-4941-v69-mkmlizer: creating bucket guanaco-reward-models
nousresearch-meta-llama-4941-v69-mkmlizer: Bucket 's3://guanaco-reward-models/' created
nousresearch-meta-llama-4941-v69-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/nousresearch-meta-llama-4941-v69_reward
nousresearch-meta-llama-4941-v69-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v69_reward/config.json
nousresearch-meta-llama-4941-v69-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v69_reward/special_tokens_map.json
nousresearch-meta-llama-4941-v69-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v69_reward/tokenizer_config.json
nousresearch-meta-llama-4941-v69-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/nousresearch-meta-llama-4941-v69_reward/merges.txt
nousresearch-meta-llama-4941-v69-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v69_reward/vocab.json
nousresearch-meta-llama-4941-v69-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v69_reward/tokenizer.json
nousresearch-meta-llama-4941-v69-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/nousresearch-meta-llama-4941-v69_reward/reward.tensors
Job nousresearch-meta-llama-4941-v69-mkmlizer completed after 94.26s with status: succeeded
Stopping job with name nousresearch-meta-llama-4941-v69-mkmlizer
Pipeline stage MKMLizer completed in 95.24s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.12s
Running pipeline stage ISVCDeployer
Creating inference service nousresearch-meta-llama-4941-v69
Waiting for inference service nousresearch-meta-llama-4941-v69 to be ready
Inference service nousresearch-meta-llama-4941-v69 ready after 40.24302935600281s
Pipeline stage ISVCDeployer completed in 47.18s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.9226484298706055s
Received healthy response to inference request in 0.3462693691253662s
Received healthy response to inference request in 1.0797886848449707s
Received healthy response to inference request in 0.859520673751831s
Received healthy response to inference request in 0.7934961318969727s
5 requests
0 failed requests
5th percentile: 0.4357147216796875
10th percentile: 0.5251600742340088
20th percentile: 0.7040507793426514
30th percentile: 0.8067010402679443
40th percentile: 0.8331108570098877
50th percentile: 0.859520673751831
60th percentile: 0.9476278781890869
70th percentile: 1.0357350826263427
80th percentile: 1.2483606338500979
90th percentile: 1.5855045318603516
95th percentile: 1.7540764808654783
99th percentile: 1.88893404006958
mean time: 1.0003446578979491
Pipeline stage StressChecker completed in 5.78s
nousresearch-meta-llama_4941_v69 status is now deployed due to DeploymentManager action
nousresearch-meta-llama_4941_v69 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of nousresearch-meta-llama_4941_v69
Running pipeline stage ISVCDeleter
Checking if service nousresearch-meta-llama-4941-v69 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 4.54s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key nousresearch-meta-llama-4941-v69/config.json from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v69/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v69/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v69/tokenizer.json from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v69/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key nousresearch-meta-llama-4941-v69_reward/config.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v69_reward/merges.txt from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v69_reward/reward.tensors from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v69_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v69_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v69_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v69_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.45s
nousresearch-meta-llama_4941_v69 status is now torndown due to DeploymentManager action