submission_id: nousresearch-meta-llama_4941_v70
developer_uid: robert_irvine
status: inactive
model_repo: NousResearch/Meta-Llama-3-8B-Instruct
reward_repo: rirv938/reward_gpt2_medium_preference_24m_e2
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': 2, '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: {'memory_template': 'Memory: {memory}\n', 'prompt_template': '{prompt}\n', 'bot_template': 'Bot: {message}\n', 'user_template': 'User: {message}\n', 'response_template': 'Bot:', 'truncate_by_message': False}
timestamp: 2024-07-03T18:20:01+00:00
model_name: nousresearch-meta-llama_4941_v70
model_group: NousResearch/Meta-Llama-
num_battles: 14218
num_wins: 6322
celo_rating: 1132.69
propriety_score: 0.7455058683702273
propriety_total_count: 6731.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 2
max_input_tokens: 512
max_output_tokens: 64
display_name: nousresearch-meta-llama_4941_v70
ineligible_reason: None
language_model: NousResearch/Meta-Llama-3-8B-Instruct
model_size: 8B
reward_model: rirv938/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-07-03
win_ratio: 0.4446476297650865
Resubmit model
Running pipeline stage MKMLizer
Starting job with name nousresearch-meta-llama-4941-v70-mkmlizer
Waiting for job on nousresearch-meta-llama-4941-v70-mkmlizer to finish
nousresearch-meta-llama-4941-v70-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
nousresearch-meta-llama-4941-v70-mkmlizer: ║ _____ __ __ ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ /___/ ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ Version: 0.8.14 ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ https://mk1.ai ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ The license key for the current software has been verified as ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ belonging to: ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ Chai Research Corp. ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
nousresearch-meta-llama-4941-v70-mkmlizer: ║ ║
nousresearch-meta-llama-4941-v70-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
nousresearch-meta-llama-4941-v70-mkmlizer: Downloaded to shared memory in 20.362s
nousresearch-meta-llama-4941-v70-mkmlizer: quantizing model to /dev/shm/model_cache
nousresearch-meta-llama-4941-v70-mkmlizer: Saving flywheel model at /dev/shm/model_cache
nousresearch-meta-llama-4941-v70-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 3/291 [00:00<00:11, 24.44it/s] Loading 0: 4%|▍ | 12/291 [00:00<00:05, 55.45it/s] Loading 0: 8%|▊ | 22/291 [00:00<00:03, 73.59it/s] Loading 0: 11%|█ | 32/291 [00:00<00:03, 82.53it/s] Loading 0: 16%|█▋ | 48/291 [00:00<00:02, 102.65it/s] Loading 0: 20%|██ | 59/291 [00:00<00:02, 96.90it/s] Loading 0: 26%|██▌ | 75/291 [00:00<00:01, 109.13it/s] Loading 0: 30%|██▉ | 86/291 [00:01<00:03, 56.09it/s] Loading 0: 35%|███▍ | 101/291 [00:01<00:02, 72.28it/s] Loading 0: 38%|███▊ | 112/291 [00:01<00:02, 70.00it/s] Loading 0: 42%|████▏ | 122/291 [00:01<00:02, 73.00it/s] Loading 0: 47%|████▋ | 138/291 [00:01<00:01, 88.43it/s] Loading 0: 51%|█████ | 149/291 [00:01<00:01, 88.14it/s] Loading 0: 57%|█████▋ | 165/291 [00:01<00:01, 100.99it/s] Loading 0: 62%|██████▏ | 179/291 [00:02<00:01, 105.84it/s] Loading 0: 66%|██████▌ | 191/291 [00:02<00:01, 62.54it/s] Loading 0: 69%|██████▊ | 200/291 [00:02<00:01, 65.70it/s] Loading 0: 72%|███████▏ | 209/291 [00:02<00:01, 69.44it/s] Loading 0: 75%|███████▌ | 219/291 [00:02<00:00, 73.86it/s] Loading 0: 78%|███████▊ | 228/291 [00:02<00:00, 75.26it/s] Loading 0: 81%|████████▏ | 237/291 [00:03<00:00, 77.48it/s] Loading 0: 85%|████████▍ | 246/291 [00:03<00:00, 79.61it/s] Loading 0: 88%|████████▊ | 255/291 [00:03<00:00, 81.91it/s] Loading 0: 91%|█████████▏| 266/291 [00:03<00:00, 84.89it/s] Loading 0: 97%|█████████▋| 281/291 [00:03<00:00, 98.15it/s] Loading 0: 100%|██████████| 291/291 [00:09<00:00, 5.94it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
nousresearch-meta-llama-4941-v70-mkmlizer: quantized model in 26.512s
nousresearch-meta-llama-4941-v70-mkmlizer: Processed model NousResearch/Meta-Llama-3-8B-Instruct in 46.874s
nousresearch-meta-llama-4941-v70-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v70/config.json
nousresearch-meta-llama-4941-v70-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v70/tokenizer_config.json
nousresearch-meta-llama-4941-v70-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v70/special_tokens_map.json
nousresearch-meta-llama-4941-v70-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v70/tokenizer.json
nousresearch-meta-llama-4941-v70-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v70/flywheel_model.0.safetensors
nousresearch-meta-llama-4941-v70-mkmlizer: loading reward model from rirv938/reward_gpt2_medium_preference_24m_e2
nousresearch-meta-llama-4941-v70-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-v70-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v70-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-v70-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v70-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-v70-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v70-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-v70-mkmlizer: warnings.warn(
nousresearch-meta-llama-4941-v70-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-v70-mkmlizer: return self.fget.__get__(instance, owner)()
nousresearch-meta-llama-4941-v70-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
nousresearch-meta-llama-4941-v70-mkmlizer: Saving duration: 0.426s
nousresearch-meta-llama-4941-v70-mkmlizer: Processed model rirv938/reward_gpt2_medium_preference_24m_e2 in 7.744s
nousresearch-meta-llama-4941-v70-mkmlizer: creating bucket guanaco-reward-models
nousresearch-meta-llama-4941-v70-mkmlizer: Bucket 's3://guanaco-reward-models/' created
nousresearch-meta-llama-4941-v70-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/nousresearch-meta-llama-4941-v70_reward
nousresearch-meta-llama-4941-v70-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v70_reward/special_tokens_map.json
nousresearch-meta-llama-4941-v70-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v70_reward/config.json
nousresearch-meta-llama-4941-v70-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v70_reward/tokenizer_config.json
nousresearch-meta-llama-4941-v70-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/nousresearch-meta-llama-4941-v70_reward/merges.txt
nousresearch-meta-llama-4941-v70-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v70_reward/vocab.json
nousresearch-meta-llama-4941-v70-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v70_reward/tokenizer.json
nousresearch-meta-llama-4941-v70-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/nousresearch-meta-llama-4941-v70_reward/reward.tensors
Job nousresearch-meta-llama-4941-v70-mkmlizer completed after 73.16s with status: succeeded
Stopping job with name nousresearch-meta-llama-4941-v70-mkmlizer
Pipeline stage MKMLizer completed in 74.05s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.13s
Running pipeline stage ISVCDeployer
Creating inference service nousresearch-meta-llama-4941-v70
Waiting for inference service nousresearch-meta-llama-4941-v70 to be ready
Inference service nousresearch-meta-llama-4941-v70 ready after 40.24569845199585s
Pipeline stage ISVCDeployer completed in 46.92s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.913398265838623s
Received healthy response to inference request in 1.1259353160858154s
Received healthy response to inference request in 0.7672028541564941s
Received healthy response to inference request in 1.0997233390808105s
Received healthy response to inference request in 0.9696362018585205s
5 requests
0 failed requests
5th percentile: 0.8076895236968994
10th percentile: 0.8481761932373046
20th percentile: 0.9291495323181153
30th percentile: 0.9956536293029785
40th percentile: 1.0476884841918945
50th percentile: 1.0997233390808105
60th percentile: 1.1102081298828126
70th percentile: 1.1206929206848144
80th percentile: 1.2834279060363771
90th percentile: 1.5984130859375
95th percentile: 1.7559056758880613
99th percentile: 1.8818997478485107
mean time: 1.1751791954040527
Pipeline stage StressChecker completed in 6.73s
nousresearch-meta-llama_4941_v70 status is now deployed due to DeploymentManager action
nousresearch-meta-llama_4941_v70 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics