submission_id: v000000-l3-8b-ugi-dontpl_8647_v1
developer_uid: v000000
status: inactive
model_repo: v000000/L3-8B-UGI-DontPlanToEnd-test
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.95, 'top_p': 0.95, 'min_p': 0.1, 'top_k': 80, 'presence_penalty': 0.05, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], '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: 2024-06-19T16:18:07+00:00
model_name: v000000-l3-8b-ugi-dontpl_8647_v1
model_group: v000000/L3-8B-UGI-DontPl
num_battles: 41001
num_wins: 23548
celo_rating: 1214.11
propriety_score: 0.7064991641760803
propriety_total_count: 19741.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: v000000-l3-8b-ugi-dontpl_8647_v1
ineligible_reason: None
language_model: v000000/L3-8B-UGI-DontPlanToEnd-test
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-19
win_ratio: 0.5743274554279164
Resubmit model
Running pipeline stage MKMLizer
Starting job with name v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer
Waiting for job on v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer to finish
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ _____ __ __ ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ /___/ ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ Version: 0.8.14 ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ https://mk1.ai ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ The license key for the current software has been verified as ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ belonging to: ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ Chai Research Corp. ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ║ ║
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/utils/_deprecation.py:131: FutureWarning: 'list_files_info' (from 'huggingface_hub.hf_api') is deprecated and will be removed from version '0.23'. Use `list_repo_tree` and `get_paths_info` instead.
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: Downloaded to shared memory in 28.775s
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: quantizing model to /dev/shm/model_cache
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:05<12:47, 2.66s/it] Loading 0: 5%|▍ | 14/291 [00:05<01:19, 3.50it/s] Loading 0: 9%|▉ | 27/291 [00:05<00:32, 8.04it/s] Loading 0: 14%|█▍ | 41/291 [00:05<00:17, 14.34it/s] Loading 0: 19%|█▉ | 55/291 [00:05<00:10, 22.45it/s] Loading 0: 23%|██▎ | 67/291 [00:06<00:10, 22.03it/s] Loading 0: 27%|██▋ | 78/291 [00:06<00:07, 28.94it/s] Loading 0: 32%|███▏ | 94/291 [00:06<00:04, 42.03it/s] Loading 0: 36%|███▌ | 105/291 [00:06<00:03, 49.50it/s] Loading 0: 40%|███▉ | 116/291 [00:06<00:02, 58.46it/s] Loading 0: 45%|████▌ | 131/291 [00:06<00:02, 72.58it/s] Loading 0: 49%|████▉ | 143/291 [00:06<00:01, 79.59it/s] Loading 0: 54%|█████▍ | 158/291 [00:07<00:01, 90.59it/s] Loading 0: 58%|█████▊ | 170/291 [00:07<00:02, 54.25it/s] Loading 0: 62%|██████▏ | 181/291 [00:07<00:01, 62.43it/s] Loading 0: 67%|██████▋ | 194/291 [00:07<00:01, 72.78it/s] Loading 0: 70%|███████ | 204/291 [00:07<00:01, 77.93it/s] Loading 0: 76%|███████▌ | 220/291 [00:07<00:00, 93.50it/s] Loading 0: 80%|███████▉ | 232/291 [00:08<00:00, 97.67it/s] Loading 0: 85%|████████▍ | 247/291 [00:08<00:00, 110.56it/s] Loading 0: 89%|████████▉ | 260/291 [00:08<00:00, 103.43it/s] Loading 0: 93%|█████████▎| 272/291 [00:08<00:00, 54.68it/s] Loading 0: 97%|█████████▋| 283/291 [00:08<00:00, 61.88it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: quantized model in 20.849s
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: Processed model v000000/L3-8B-UGI-DontPlanToEnd-test in 51.150s
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: creating bucket guanaco-mkml-models
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/v000000-l3-8b-ugi-dontpl-8647-v1
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/v000000-l3-8b-ugi-dontpl-8647-v1/special_tokens_map.json
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/v000000-l3-8b-ugi-dontpl-8647-v1/tokenizer_config.json
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/v000000-l3-8b-ugi-dontpl-8647-v1/config.json
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/v000000-l3-8b-ugi-dontpl-8647-v1/tokenizer.json
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/v000000-l3-8b-ugi-dontpl-8647-v1/flywheel_model.0.safetensors
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:913: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: warnings.warn(
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:757: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: warnings.warn(
v000000-l3-8b-ugi-dontpl-8647-v1-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.
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: warnings.warn(
v000000-l3-8b-ugi-dontpl-8647-v1-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()
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: return self.fget.__get__(instance, owner)()
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: Saving duration: 0.257s
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.295s
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: creating bucket guanaco-reward-models
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/v000000-l3-8b-ugi-dontpl-8647-v1_reward
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/v000000-l3-8b-ugi-dontpl-8647-v1_reward/config.json
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/v000000-l3-8b-ugi-dontpl-8647-v1_reward/tokenizer_config.json
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/v000000-l3-8b-ugi-dontpl-8647-v1_reward/merges.txt
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/v000000-l3-8b-ugi-dontpl-8647-v1_reward/special_tokens_map.json
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/v000000-l3-8b-ugi-dontpl-8647-v1_reward/vocab.json
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/v000000-l3-8b-ugi-dontpl-8647-v1_reward/tokenizer.json
v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/v000000-l3-8b-ugi-dontpl-8647-v1_reward/reward.tensors
Job v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer completed after 72.97s with status: succeeded
Stopping job with name v000000-l3-8b-ugi-dontpl-8647-v1-mkmlizer
Pipeline stage MKMLizer completed in 73.95s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service v000000-l3-8b-ugi-dontpl-8647-v1
Waiting for inference service v000000-l3-8b-ugi-dontpl-8647-v1 to be ready
Inference service v000000-l3-8b-ugi-dontpl-8647-v1 ready after 50.318596839904785s
Pipeline stage ISVCDeployer completed in 56.35s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.955216884613037s
Received healthy response to inference request in 1.0741853713989258s
Received healthy response to inference request in 1.056586742401123s
Received healthy response to inference request in 1.0394554138183594s
Received healthy response to inference request in 1.0731909275054932s
5 requests
0 failed requests
5th percentile: 1.042881679534912
10th percentile: 1.0463079452514648
20th percentile: 1.0531604766845704
30th percentile: 1.059907579421997
40th percentile: 1.066549253463745
50th percentile: 1.0731909275054932
60th percentile: 1.0735887050628663
70th percentile: 1.0739864826202392
80th percentile: 1.2503916740417482
90th percentile: 1.6028042793273927
95th percentile: 1.7790105819702147
99th percentile: 1.9199756240844725
mean time: 1.2397270679473877
Pipeline stage StressChecker completed in 6.88s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.04s
v000000-l3-8b-ugi-dontpl_8647_v1 status is now deployed due to DeploymentManager action
v000000-l3-8b-ugi-dontpl_8647_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics