submission_id: chaiml-llamademo_v5
developer_uid: Selenee
status: inactive
model_repo: ChaiML/llamademo
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 100.0, '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': 5, '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: 2024-06-20T00:11:21+00:00
model_name: testBot
model_group: ChaiML/llamademo
num_battles: 13711
num_wins: 3681
celo_rating: 815.46
propriety_score: 0.6451859788764733
propriety_total_count: 6533.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 6738415616.0
best_of: 5
max_input_tokens: 512
max_output_tokens: 64
display_name: testBot
ineligible_reason: None
language_model: ChaiML/llamademo
model_size: 7B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-19
win_ratio: 0.2684705710743199
Resubmit model
Running pipeline stage MKMLizer
Starting job with name chaiml-llamademo-v5-mkmlizer
Waiting for job on chaiml-llamademo-v5-mkmlizer to finish
chaiml-llamademo-v5-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-llamademo-v5-mkmlizer: ║ _____ __ __ ║
chaiml-llamademo-v5-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
chaiml-llamademo-v5-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
chaiml-llamademo-v5-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
chaiml-llamademo-v5-mkmlizer: ║ /___/ ║
chaiml-llamademo-v5-mkmlizer: ║ ║
chaiml-llamademo-v5-mkmlizer: ║ Version: 0.8.14 ║
chaiml-llamademo-v5-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
chaiml-llamademo-v5-mkmlizer: ║ https://mk1.ai ║
chaiml-llamademo-v5-mkmlizer: ║ ║
chaiml-llamademo-v5-mkmlizer: ║ The license key for the current software has been verified as ║
chaiml-llamademo-v5-mkmlizer: ║ belonging to: ║
chaiml-llamademo-v5-mkmlizer: ║ ║
chaiml-llamademo-v5-mkmlizer: ║ Chai Research Corp. ║
chaiml-llamademo-v5-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
chaiml-llamademo-v5-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
chaiml-llamademo-v5-mkmlizer: ║ ║
chaiml-llamademo-v5-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-llamademo-v5-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.
chaiml-llamademo-v5-mkmlizer: warnings.warn(warning_message, FutureWarning)
chaiml-llamademo-v5-mkmlizer: Downloaded to shared memory in 14.910s
chaiml-llamademo-v5-mkmlizer: quantizing model to /dev/shm/model_cache
chaiml-llamademo-v5-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Failed to get response for submission neversleep-noromaid-v0-_8068_v36: ('http://neversleep-noromaid-v0-8068-v36-predictor-default.tenant-chaiml-guanaco.knative.ord1.coreweave.cloud/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:43216->127.0.0.1:8080: read: connection reset by peer\n')
chaiml-llamademo-v5-mkmlizer: quantized model in 13.966s
chaiml-llamademo-v5-mkmlizer: Processed model ChaiML/llamademo in 31.118s
chaiml-llamademo-v5-mkmlizer: creating bucket guanaco-mkml-models
chaiml-llamademo-v5-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-llamademo-v5-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-llamademo-v5
chaiml-llamademo-v5-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-llamademo-v5/config.json
chaiml-llamademo-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/chaiml-llamademo-v5/tokenizer.model
chaiml-llamademo-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-llamademo-v5/tokenizer_config.json
chaiml-llamademo-v5-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-llamademo-v5/special_tokens_map.json
chaiml-llamademo-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-llamademo-v5/tokenizer.json
chaiml-llamademo-v5-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-llamademo-v5/flywheel_model.0.safetensors
chaiml-llamademo-v5-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
chaiml-llamademo-v5-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 73%|███████▎ | 213/291 [00:01<00:00, 203.72it/s] Loading 0: 100%|██████████| 291/291 [00:02<00:00, 113.92it/s] /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.
chaiml-llamademo-v5-mkmlizer: warnings.warn(
chaiml-llamademo-v5-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.
chaiml-llamademo-v5-mkmlizer: warnings.warn(
chaiml-llamademo-v5-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.
chaiml-llamademo-v5-mkmlizer: warnings.warn(
chaiml-llamademo-v5-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()
chaiml-llamademo-v5-mkmlizer: return self.fget.__get__(instance, owner)()
chaiml-llamademo-v5-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
chaiml-llamademo-v5-mkmlizer: Saving duration: 0.420s
chaiml-llamademo-v5-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 7.384s
chaiml-llamademo-v5-mkmlizer: creating bucket guanaco-reward-models
chaiml-llamademo-v5-mkmlizer: Bucket 's3://guanaco-reward-models/' created
chaiml-llamademo-v5-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/chaiml-llamademo-v5_reward
chaiml-llamademo-v5-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/chaiml-llamademo-v5_reward/config.json
chaiml-llamademo-v5-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/chaiml-llamademo-v5_reward/tokenizer_config.json
chaiml-llamademo-v5-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/chaiml-llamademo-v5_reward/special_tokens_map.json
chaiml-llamademo-v5-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/chaiml-llamademo-v5_reward/merges.txt
chaiml-llamademo-v5-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/chaiml-llamademo-v5_reward/vocab.json
chaiml-llamademo-v5-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/chaiml-llamademo-v5_reward/tokenizer.json
chaiml-llamademo-v5-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/chaiml-llamademo-v5_reward/reward.tensors
Job chaiml-llamademo-v5-mkmlizer completed after 62.95s with status: succeeded
Stopping job with name chaiml-llamademo-v5-mkmlizer
Pipeline stage MKMLizer completed in 66.01s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service chaiml-llamademo-v5
Waiting for inference service chaiml-llamademo-v5 to be ready
Inference service chaiml-llamademo-v5 ready after 90.5761227607727s
Pipeline stage ISVCDeployer completed in 97.53s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.9740476608276367s
Received healthy response to inference request in 0.9785659313201904s
Received healthy response to inference request in 0.4647531509399414s
Received healthy response to inference request in 1.0096876621246338s
Received healthy response to inference request in 1.0069589614868164s
5 requests
0 failed requests
5th percentile: 0.5675157070159912
10th percentile: 0.670278263092041
20th percentile: 0.8758033752441406
30th percentile: 0.9842445373535156
40th percentile: 0.995601749420166
50th percentile: 1.0069589614868164
60th percentile: 1.0080504417419434
70th percentile: 1.0091419219970703
80th percentile: 1.2025596618652346
90th percentile: 1.5883036613464356
95th percentile: 1.781175661087036
99th percentile: 1.9354732608795167
mean time: 1.0868026733398437
Pipeline stage StressChecker completed in 6.21s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.04s
chaiml-llamademo_v5 status is now deployed due to DeploymentManager action
chaiml-llamademo_v5 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics