submission_id: meta-llama-meta-llama-3-8b_v4
developer_uid: Meliodia
status: inactive
model_repo: meta-llama/Meta-Llama-3-8B
reward_repo: ChaiML/gpt2_medium_pairwise_60m_step_937500
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': ['\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: {'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}
timestamp: 2024-07-02T16:48:53+00:00
model_name: meta-base-model
model_group: meta-llama/Meta-Llama-3-
num_battles: 12290
num_wins: 5534
celo_rating: 1138.99
propriety_score: 0.7177476835352815
propriety_total_count: 5612.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 4
max_input_tokens: 512
max_output_tokens: 64
display_name: meta-base-model
ineligible_reason: None
language_model: meta-llama/Meta-Llama-3-8B
model_size: 8B
reward_model: ChaiML/gpt2_medium_pairwise_60m_step_937500
us_pacific_date: 2024-07-02
win_ratio: 0.4502847843775427
Resubmit model
Running pipeline stage MKMLizer
Starting job with name meta-llama-meta-llama-3-8b-v4-mkmlizer
Waiting for job on meta-llama-meta-llama-3-8b-v4-mkmlizer to finish
meta-llama-meta-llama-3-8b-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ _____ __ __ ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ /___/ ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ Version: 0.8.14 ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ https://mk1.ai ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ The license key for the current software has been verified as ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ belonging to: ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ Chai Research Corp. ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ║ ║
meta-llama-meta-llama-3-8b-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
meta-llama-meta-llama-3-8b-v4-mkmlizer: Downloaded to shared memory in 40.266s
meta-llama-meta-llama-3-8b-v4-mkmlizer: quantizing model to /dev/shm/model_cache
meta-llama-meta-llama-3-8b-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
meta-llama-meta-llama-3-8b-v4-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 3/291 [00:02<04:46, 1.00it/s] Loading 0: 1%|▏ | 4/291 [00:03<04:13, 1.13it/s] Loading 0: 2%|▏ | 5/291 [00:04<04:28, 1.06it/s] Loading 0: 7%|▋ | 21/291 [00:04<00:34, 7.84it/s] Loading 0: 8%|▊ | 24/291 [00:06<00:51, 5.17it/s] Loading 0: 9%|▉ | 26/291 [00:06<00:53, 4.97it/s] Loading 0: 10%|▉ | 28/291 [00:07<00:55, 4.77it/s] Loading 0: 10%|█ | 30/291 [00:08<01:19, 3.28it/s] Loading 0: 11%|█ | 31/291 [00:09<01:47, 2.42it/s] Loading 0: 11%|█ | 32/291 [00:11<02:20, 1.84it/s] Loading 0: 12%|█▏ | 35/291 [00:11<01:42, 2.50it/s] Loading 0: 12%|█▏ | 36/291 [00:12<01:41, 2.51it/s] Loading 0: 13%|█▎ | 37/291 [00:12<01:29, 2.85it/s] Loading 0: 13%|█▎ | 39/291 [00:13<01:50, 2.29it/s] Loading 0: 14%|█▎ | 40/291 [00:14<02:11, 1.91it/s] Loading 0: 14%|█▍ | 41/291 [00:14<02:12, 1.88it/s] Loading 0: 15%|█▌ | 44/291 [00:14<01:12, 3.42it/s] Loading 0: 16%|█▌ | 46/291 [00:15<00:52, 4.64it/s] Loading 0: 20%|█▉ | 58/291 [00:15<00:15, 15.52it/s] Loading 0: 25%|██▌ | 73/291 [00:15<00:06, 31.47it/s] Loading 0: 29%|██▊ | 83/291 [00:15<00:06, 30.97it/s] Loading 0: 32%|███▏ | 94/291 [00:15<00:04, 41.29it/s] Loading 0: 37%|███▋ | 108/291 [00:15<00:03, 56.66it/s] Loading 0: 42%|████▏ | 121/291 [00:15<00:02, 69.47it/s] Loading 0: 47%|████▋ | 138/291 [00:16<00:01, 87.23it/s] Loading 0: 52%|█████▏ | 150/291 [00:16<00:01, 93.65it/s] Loading 0: 57%|█████▋ | 166/291 [00:16<00:01, 105.81it/s] Loading 0: 62%|██████▏ | 180/291 [00:16<00:00, 114.26it/s] Loading 0: 66%|██████▋ | 193/291 [00:16<00:01, 71.53it/s] Loading 0: 72%|███████▏ | 209/291 [00:16<00:00, 87.82it/s] Loading 0: 76%|███████▌ | 221/291 [00:16<00:00, 90.74it/s] Loading 0: 82%|████████▏ | 238/291 [00:17<00:00, 104.35it/s] Loading 0: 88%|████████▊ | 255/291 [00:17<00:00, 115.32it/s] Loading 0: 92%|█████████▏| 268/291 [00:17<00:00, 116.88it/s] Loading 0: 97%|█████████▋| 282/291 [00:17<00:00, 118.73it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
meta-llama-meta-llama-3-8b-v4-mkmlizer: quantized model in 51.808s
meta-llama-meta-llama-3-8b-v4-mkmlizer: Processed model meta-llama/Meta-Llama-3-8B in 92.074s
meta-llama-meta-llama-3-8b-v4-mkmlizer: creating bucket guanaco-mkml-models
meta-llama-meta-llama-3-8b-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
meta-llama-meta-llama-3-8b-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/meta-llama-meta-llama-3-8b-v4
meta-llama-meta-llama-3-8b-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/meta-llama-meta-llama-3-8b-v4/config.json
meta-llama-meta-llama-3-8b-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/meta-llama-meta-llama-3-8b-v4/tokenizer_config.json
meta-llama-meta-llama-3-8b-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/meta-llama-meta-llama-3-8b-v4/special_tokens_map.json
meta-llama-meta-llama-3-8b-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/meta-llama-meta-llama-3-8b-v4/tokenizer.json
meta-llama-meta-llama-3-8b-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/meta-llama-meta-llama-3-8b-v4/flywheel_model.0.safetensors
meta-llama-meta-llama-3-8b-v4-mkmlizer: loading reward model from ChaiML/gpt2_medium_pairwise_60m_step_937500
meta-llama-meta-llama-3-8b-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.
meta-llama-meta-llama-3-8b-v4-mkmlizer: warnings.warn(
meta-llama-meta-llama-3-8b-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`.
meta-llama-meta-llama-3-8b-v4-mkmlizer: warnings.warn(
meta-llama-meta-llama-3-8b-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.
meta-llama-meta-llama-3-8b-v4-mkmlizer: warnings.warn(
meta-llama-meta-llama-3-8b-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.
meta-llama-meta-llama-3-8b-v4-mkmlizer: warnings.warn(
meta-llama-meta-llama-3-8b-v4-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
meta-llama-meta-llama-3-8b-v4-mkmlizer: Saving duration: 0.396s
meta-llama-meta-llama-3-8b-v4-mkmlizer: Processed model ChaiML/gpt2_medium_pairwise_60m_step_937500 in 3.997s
meta-llama-meta-llama-3-8b-v4-mkmlizer: creating bucket guanaco-reward-models
meta-llama-meta-llama-3-8b-v4-mkmlizer: Bucket 's3://guanaco-reward-models/' created
meta-llama-meta-llama-3-8b-v4-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/meta-llama-meta-llama-3-8b-v4_reward
meta-llama-meta-llama-3-8b-v4-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/meta-llama-meta-llama-3-8b-v4_reward/special_tokens_map.json
meta-llama-meta-llama-3-8b-v4-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/meta-llama-meta-llama-3-8b-v4_reward/merges.txt
meta-llama-meta-llama-3-8b-v4-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/meta-llama-meta-llama-3-8b-v4_reward/config.json
meta-llama-meta-llama-3-8b-v4-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/meta-llama-meta-llama-3-8b-v4_reward/vocab.json
meta-llama-meta-llama-3-8b-v4-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/meta-llama-meta-llama-3-8b-v4_reward/tokenizer_config.json
meta-llama-meta-llama-3-8b-v4-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/meta-llama-meta-llama-3-8b-v4_reward/tokenizer.json
meta-llama-meta-llama-3-8b-v4-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/meta-llama-meta-llama-3-8b-v4_reward/reward.tensors
Job meta-llama-meta-llama-3-8b-v4-mkmlizer completed after 115.95s with status: succeeded
Stopping job with name meta-llama-meta-llama-3-8b-v4-mkmlizer
Pipeline stage MKMLizer completed in 117.03s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service meta-llama-meta-llama-3-8b-v4
Waiting for inference service meta-llama-meta-llama-3-8b-v4 to be ready
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Inference service meta-llama-meta-llama-3-8b-v4 ready after 40.342358350753784s
Pipeline stage ISVCDeployer completed in 47.59s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.8045780658721924s
Received healthy response to inference request in 1.1126117706298828s
Received healthy response to inference request in 1.1116480827331543s
Received healthy response to inference request in 0.8134405612945557s
Received healthy response to inference request in 1.1664752960205078s
5 requests
0 failed requests
5th percentile: 0.8730820655822754
10th percentile: 0.9327235698699952
20th percentile: 1.0520065784454347
30th percentile: 1.1118408203125
40th percentile: 1.1122262954711915
50th percentile: 1.1126117706298828
60th percentile: 1.1341571807861328
70th percentile: 1.1557025909423828
80th percentile: 1.2940958499908448
90th percentile: 1.5493369579315186
95th percentile: 1.6769575119018554
99th percentile: 1.779053955078125
mean time: 1.2017507553100586
Pipeline stage StressChecker completed in 6.75s
meta-llama-meta-llama-3-8b_v4 status is now deployed due to DeploymentManager action
meta-llama-meta-llama-3-8b_v4 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics