submission_id: meseca-sophon-1_v3
developer_uid: nguyenzzz
status: inactive
model_repo: meseca/sophon-1
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 50, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 16, '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: 2024-06-02T11:00:40+00:00
model_name: meseca-sophon-1_v1
model_eval_status: success
model_group: meseca/sophon-1
num_battles: 11681
num_wins: 6235
celo_rating: 1200.2
safety_score: 0.94
propriety_score: 0.6205533596837944
propriety_total_count: 253.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: meseca-sophon-1_v1
ineligible_reason: propriety_total_count < 5000
language_model: meseca/sophon-1
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-02
win_ratio: 0.5337727934252204
Resubmit model
Running pipeline stage MKMLizer
Starting job with name meseca-sophon-1-v3-mkmlizer
Waiting for job on meseca-sophon-1-v3-mkmlizer to finish
meseca-sophon-1-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
meseca-sophon-1-v3-mkmlizer: ║ _____ __ __ ║
meseca-sophon-1-v3-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
meseca-sophon-1-v3-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
meseca-sophon-1-v3-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
meseca-sophon-1-v3-mkmlizer: ║ /___/ ║
meseca-sophon-1-v3-mkmlizer: ║ ║
meseca-sophon-1-v3-mkmlizer: ║ Version: 0.8.14 ║
meseca-sophon-1-v3-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
meseca-sophon-1-v3-mkmlizer: ║ https://mk1.ai ║
meseca-sophon-1-v3-mkmlizer: ║ ║
meseca-sophon-1-v3-mkmlizer: ║ The license key for the current software has been verified as ║
meseca-sophon-1-v3-mkmlizer: ║ belonging to: ║
meseca-sophon-1-v3-mkmlizer: ║ ║
meseca-sophon-1-v3-mkmlizer: ║ Chai Research Corp. ║
meseca-sophon-1-v3-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
meseca-sophon-1-v3-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
meseca-sophon-1-v3-mkmlizer: ║ ║
meseca-sophon-1-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
meseca-sophon-1-v3-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.
meseca-sophon-1-v3-mkmlizer: warnings.warn(warning_message, FutureWarning)
meseca-sophon-1-v3-mkmlizer: Downloaded to shared memory in 28.174s
meseca-sophon-1-v3-mkmlizer: quantizing model to /dev/shm/model_cache
meseca-sophon-1-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
meseca-sophon-1-v3-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 4%|▍ | 13/291 [00:00<00:02, 124.67it/s] Loading 0: 11%|█ | 31/291 [00:00<00:01, 146.08it/s] Loading 0: 16%|█▌ | 46/291 [00:00<00:02, 93.39it/s] Loading 0: 22%|██▏ | 64/291 [00:00<00:01, 115.69it/s] Loading 0: 27%|██▋ | 80/291 [00:00<00:02, 97.74it/s] Loading 0: 32%|███▏ | 92/291 [00:00<00:01, 101.29it/s] Loading 0: 40%|███▉ | 115/291 [00:00<00:01, 131.98it/s] Loading 0: 45%|████▍ | 130/291 [00:10<00:31, 5.11it/s] Loading 0: 49%|████▉ | 144/291 [00:11<00:21, 6.84it/s] Loading 0: 54%|█████▍ | 158/291 [00:11<00:14, 9.33it/s] Loading 0: 60%|██████ | 176/291 [00:11<00:08, 13.78it/s] Loading 0: 65%|██████▍ | 189/291 [00:11<00:05, 17.25it/s] Loading 0: 70%|███████ | 205/291 [00:11<00:03, 23.88it/s] Loading 0: 76%|███████▌ | 220/291 [00:11<00:02, 29.78it/s] Loading 0: 82%|████████▏ | 239/291 [00:12<00:01, 41.89it/s] Loading 0: 88%|████████▊ | 257/291 [00:12<00:00, 54.97it/s] Loading 0: 93%|█████████▎| 271/291 [00:12<00:00, 56.66it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
meseca-sophon-1-v3-mkmlizer: quantized model in 23.061s
meseca-sophon-1-v3-mkmlizer: Processed model meseca/sophon-1 in 53.738s
meseca-sophon-1-v3-mkmlizer: creating bucket guanaco-mkml-models
meseca-sophon-1-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
meseca-sophon-1-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/meseca-sophon-1-v3
meseca-sophon-1-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/meseca-sophon-1-v3/tokenizer_config.json
meseca-sophon-1-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/meseca-sophon-1-v3/special_tokens_map.json
meseca-sophon-1-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/meseca-sophon-1-v3/tokenizer.json
meseca-sophon-1-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/meseca-sophon-1-v3/config.json
meseca-sophon-1-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/meseca-sophon-1-v3/flywheel_model.0.safetensors
meseca-sophon-1-v3-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
meseca-sophon-1-v3-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.
meseca-sophon-1-v3-mkmlizer: warnings.warn(
meseca-sophon-1-v3-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.
meseca-sophon-1-v3-mkmlizer: warnings.warn(
meseca-sophon-1-v3-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.
meseca-sophon-1-v3-mkmlizer: warnings.warn(
meseca-sophon-1-v3-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()
meseca-sophon-1-v3-mkmlizer: return self.fget.__get__(instance, owner)()
meseca-sophon-1-v3-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
meseca-sophon-1-v3-mkmlizer: Saving duration: 0.414s
meseca-sophon-1-v3-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.042s
meseca-sophon-1-v3-mkmlizer: creating bucket guanaco-reward-models
meseca-sophon-1-v3-mkmlizer: Bucket 's3://guanaco-reward-models/' created
meseca-sophon-1-v3-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/meseca-sophon-1-v3_reward
meseca-sophon-1-v3-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/meseca-sophon-1-v3_reward/config.json
meseca-sophon-1-v3-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/meseca-sophon-1-v3_reward/tokenizer_config.json
meseca-sophon-1-v3-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/meseca-sophon-1-v3_reward/special_tokens_map.json
meseca-sophon-1-v3-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/meseca-sophon-1-v3_reward/merges.txt
meseca-sophon-1-v3-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/meseca-sophon-1-v3_reward/vocab.json
meseca-sophon-1-v3-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/meseca-sophon-1-v3_reward/tokenizer.json
meseca-sophon-1-v3-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/meseca-sophon-1-v3_reward/reward.tensors
Job meseca-sophon-1-v3-mkmlizer completed after 73.29s with status: succeeded
Stopping job with name meseca-sophon-1-v3-mkmlizer
Pipeline stage MKMLizer completed in 73.64s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service meseca-sophon-1-v3
Waiting for inference service meseca-sophon-1-v3 to be ready
Inference service meseca-sophon-1-v3 ready after 50.579906940460205s
Pipeline stage ISVCDeployer completed in 56.16s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.314635992050171s
Received healthy response to inference request in 1.253739356994629s
Received healthy response to inference request in 1.241330862045288s
Received healthy response to inference request in 1.2562708854675293s
Received healthy response to inference request in 1.2613003253936768s
5 requests
0 failed requests
5th percentile: 1.2438125610351562
10th percentile: 1.2462942600250244
20th percentile: 1.2512576580047607
30th percentile: 1.254245662689209
40th percentile: 1.255258274078369
50th percentile: 1.2562708854675293
60th percentile: 1.2582826614379883
70th percentile: 1.2602944374084473
80th percentile: 1.4719674587249758
90th percentile: 1.8933017253875732
95th percentile: 2.103968858718872
99th percentile: 2.272502565383911
mean time: 1.4654554843902587
Pipeline stage StressChecker completed in 7.97s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running pipeline stage DaemonicSafetyScorer
Running M-Eval for topic stay_in_character
Pipeline stage DaemonicSafetyScorer completed in 0.04s
M-Eval Dataset for topic stay_in_character is loaded
meseca-sophon-1_v3 status is now deployed due to DeploymentManager action
meseca-sophon-1_v3 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics