submission_id: sao10k-l3-rp-v5-2_v6
developer_uid: sao10k
status: inactive
model_repo: Sao10K/L3-RP-v5.2
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.2, 'top_p': 1.0, 'min_p': 0.1, '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: {'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: 2024-07-11T14:16:27+00:00
model_name: 5-2-2
model_group: Sao10K/L3-RP-v5.2
num_battles: 44816
num_wins: 23906
celo_rating: 1217.55
propriety_score: 0.7109723461195361
propriety_total_count: 7847.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: 5-2-2
ineligible_reason: None
language_model: Sao10K/L3-RP-v5.2
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-07-11
win_ratio: 0.5334255622991788
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name sao10k-l3-rp-v5-2-v6-mkmlizer
Waiting for job on sao10k-l3-rp-v5-2-v6-mkmlizer to finish
sao10k-l3-rp-v5-2-v6-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ _____ __ __ ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ /___/ ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ Version: 0.8.14 ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ https://mk1.ai ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ The license key for the current software has been verified as ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ belonging to: ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ Chai Research Corp. ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ║ ║
sao10k-l3-rp-v5-2-v6-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
sao10k-l3-rp-v5-2-v6-mkmlizer: Downloaded to shared memory in 63.177s
sao10k-l3-rp-v5-2-v6-mkmlizer: quantizing model to /dev/shm/model_cache
sao10k-l3-rp-v5-2-v6-mkmlizer: Saving flywheel model at /dev/shm/model_cache
sao10k-l3-rp-v5-2-v6-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:05<13:40, 2.84s/it] Loading 0: 5%|▍ | 14/291 [00:05<01:24, 3.28it/s] Loading 0: 10%|▉ | 28/291 [00:05<00:33, 7.88it/s] Loading 0: 14%|█▍ | 41/291 [00:06<00:18, 13.40it/s] Loading 0: 19%|█▊ | 54/291 [00:06<00:11, 20.50it/s] Loading 0: 23%|██▎ | 66/291 [00:06<00:10, 21.71it/s] Loading 0: 26%|██▋ | 77/291 [00:06<00:07, 28.52it/s] Loading 0: 31%|███▏ | 91/291 [00:06<00:05, 39.68it/s] Loading 0: 36%|███▌ | 104/291 [00:06<00:03, 50.30it/s] Loading 0: 40%|████ | 117/291 [00:07<00:02, 62.20it/s] Loading 0: 45%|████▍ | 130/291 [00:07<00:02, 73.94it/s] Loading 0: 49%|████▉ | 142/291 [00:07<00:01, 80.07it/s] Loading 0: 54%|█████▍ | 157/291 [00:07<00:01, 94.31it/s] Loading 0: 58%|█████▊ | 170/291 [00:07<00:02, 53.98it/s] Loading 0: 63%|██████▎ | 184/291 [00:07<00:01, 66.71it/s] Loading 0: 67%|██████▋ | 195/291 [00:08<00:01, 73.57it/s] Loading 0: 73%|███████▎ | 211/291 [00:08<00:00, 88.35it/s] Loading 0: 77%|███████▋ | 223/291 [00:08<00:00, 92.49it/s] Loading 0: 82%|████████▏ | 238/291 [00:08<00:00, 105.08it/s] Loading 0: 86%|████████▋ | 251/291 [00:08<00:00, 102.84it/s] Loading 0: 91%|█████████ | 265/291 [00:08<00:00, 111.01it/s] Loading 0: 96%|█████████▌| 278/291 [00:09<00:00, 59.08it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
sao10k-l3-rp-v5-2-v6-mkmlizer: quantized model in 29.779s
sao10k-l3-rp-v5-2-v6-mkmlizer: Processed model Sao10K/L3-RP-v5.2 in 92.956s
sao10k-l3-rp-v5-2-v6-mkmlizer: creating bucket guanaco-mkml-models
sao10k-l3-rp-v5-2-v6-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
sao10k-l3-rp-v5-2-v6-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/sao10k-l3-rp-v5-2-v6
sao10k-l3-rp-v5-2-v6-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/sao10k-l3-rp-v5-2-v6/special_tokens_map.json
sao10k-l3-rp-v5-2-v6-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/sao10k-l3-rp-v5-2-v6/config.json
sao10k-l3-rp-v5-2-v6-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/sao10k-l3-rp-v5-2-v6/tokenizer_config.json
sao10k-l3-rp-v5-2-v6-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/sao10k-l3-rp-v5-2-v6/tokenizer.json
sao10k-l3-rp-v5-2-v6-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/sao10k-l3-rp-v5-2-v6/flywheel_model.0.safetensors
sao10k-l3-rp-v5-2-v6-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
sao10k-l3-rp-v5-2-v6-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.
sao10k-l3-rp-v5-2-v6-mkmlizer: warnings.warn(
sao10k-l3-rp-v5-2-v6-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`.
sao10k-l3-rp-v5-2-v6-mkmlizer: warnings.warn(
sao10k-l3-rp-v5-2-v6-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.
sao10k-l3-rp-v5-2-v6-mkmlizer: warnings.warn(
sao10k-l3-rp-v5-2-v6-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.
sao10k-l3-rp-v5-2-v6-mkmlizer: warnings.warn(
sao10k-l3-rp-v5-2-v6-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()
sao10k-l3-rp-v5-2-v6-mkmlizer: return self.fget.__get__(instance, owner)()
sao10k-l3-rp-v5-2-v6-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
sao10k-l3-rp-v5-2-v6-mkmlizer: Saving duration: 0.534s
sao10k-l3-rp-v5-2-v6-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 8.501s
sao10k-l3-rp-v5-2-v6-mkmlizer: creating bucket guanaco-reward-models
sao10k-l3-rp-v5-2-v6-mkmlizer: Bucket 's3://guanaco-reward-models/' created
sao10k-l3-rp-v5-2-v6-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/sao10k-l3-rp-v5-2-v6_reward
sao10k-l3-rp-v5-2-v6-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/sao10k-l3-rp-v5-2-v6_reward/config.json
sao10k-l3-rp-v5-2-v6-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/sao10k-l3-rp-v5-2-v6_reward/special_tokens_map.json
sao10k-l3-rp-v5-2-v6-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/sao10k-l3-rp-v5-2-v6_reward/vocab.json
sao10k-l3-rp-v5-2-v6-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/sao10k-l3-rp-v5-2-v6_reward/tokenizer_config.json
sao10k-l3-rp-v5-2-v6-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/sao10k-l3-rp-v5-2-v6_reward/merges.txt
sao10k-l3-rp-v5-2-v6-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/sao10k-l3-rp-v5-2-v6_reward/tokenizer.json
sao10k-l3-rp-v5-2-v6-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/sao10k-l3-rp-v5-2-v6_reward/reward.tensors
Job sao10k-l3-rp-v5-2-v6-mkmlizer completed after 124.6s with status: succeeded
Stopping job with name sao10k-l3-rp-v5-2-v6-mkmlizer
Pipeline stage MKMLizer completed in 125.54s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service sao10k-l3-rp-v5-2-v6
Waiting for inference service sao10k-l3-rp-v5-2-v6 to be ready
Inference service sao10k-l3-rp-v5-2-v6 ready after 40.22998404502869s
Pipeline stage ISVCDeployer completed in 47.02s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.048549175262451s
Received healthy response to inference request in 1.3464372158050537s
Received healthy response to inference request in 1.3416860103607178s
Received healthy response to inference request in 1.2928483486175537s
Received healthy response to inference request in 1.3522346019744873s
5 requests
0 failed requests
5th percentile: 1.3026158809661865
10th percentile: 1.3123834133148193
20th percentile: 1.331918478012085
30th percentile: 1.3426362514495849
40th percentile: 1.3445367336273193
50th percentile: 1.3464372158050537
60th percentile: 1.3487561702728272
70th percentile: 1.3510751247406005
80th percentile: 1.49149751663208
90th percentile: 1.7700233459472656
95th percentile: 1.9092862606048584
99th percentile: 2.020696592330933
mean time: 1.4763510704040528
Pipeline stage StressChecker completed in 8.08s
sao10k-l3-rp-v5-2_v6 status is now deployed due to DeploymentManager action
sao10k-l3-rp-v5-2_v6 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics