submission_id: cycy233-misllm-v4-3_v1
developer_uid: shiroe40
status: inactive
model_repo: cycy233/misllm_v4.3
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.9, 'top_p': 0.8, 'min_p': 0.0, 'top_k': 40, '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': True}
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': True}
timestamp: 2024-05-20T14:44:33+00:00
model_name: base
model_eval_status: success
model_group: cycy233/misllm_v4.3
num_battles: 17156
num_wins: 8899
celo_rating: 1183.01
safety_score: 0.94
propriety_score: 0.0
propriety_total_count: 0.0
submission_type: basic
model_architecture: MistralForCausalLM
model_num_parameters: 7241732096.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: base
ineligible_reason: propriety_total_count < 5000
language_model: cycy233/misllm_v4.3
model_size: 7B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-20
win_ratio: 0.518710655164374
Resubmit model
Running pipeline stage MKMLizer
Starting job with name cycy233-misllm-v4-3-v1-mkmlizer
Waiting for job on cycy233-misllm-v4-3-v1-mkmlizer to finish
cycy233-misllm-v4-3-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
cycy233-misllm-v4-3-v1-mkmlizer: ║ _____ __ __ ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ /___/ ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ Version: 0.8.14 ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ https://mk1.ai ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ The license key for the current software has been verified as ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ belonging to: ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ Chai Research Corp. ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
cycy233-misllm-v4-3-v1-mkmlizer: ║ ║
cycy233-misllm-v4-3-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
cycy233-misllm-v4-3-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.
cycy233-misllm-v4-3-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
cycy233-misllm-v4-3-v1-mkmlizer: Downloaded to shared memory in 24.634s
cycy233-misllm-v4-3-v1-mkmlizer: quantizing model to /dev/shm/model_cache
cycy233-misllm-v4-3-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
cycy233-misllm-v4-3-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 34%|███▎ | 98/291 [00:00<00:01, 99.21it/s] Loading 0: 70%|███████ | 204/291 [00:01<00:00, 111.10it/s] Loading 0: 100%|██████████| 291/291 [00:03<00:00, 87.00it/s] You set `add_prefix_space`. The tokenizer needs to be converted from the slow tokenizers
cycy233-misllm-v4-3-v1-mkmlizer: quantized model in 14.043s
cycy233-misllm-v4-3-v1-mkmlizer: Processed model cycy233/misllm_v4.3 in 40.980s
cycy233-misllm-v4-3-v1-mkmlizer: creating bucket guanaco-mkml-models
cycy233-misllm-v4-3-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
cycy233-misllm-v4-3-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/cycy233-misllm-v4-3-v1
cycy233-misllm-v4-3-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/cycy233-misllm-v4-3-v1/config.json
cycy233-misllm-v4-3-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/cycy233-misllm-v4-3-v1/tokenizer_config.json
cycy233-misllm-v4-3-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/cycy233-misllm-v4-3-v1/tokenizer.model
cycy233-misllm-v4-3-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/cycy233-misllm-v4-3-v1/tokenizer.json
cycy233-misllm-v4-3-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/cycy233-misllm-v4-3-v1/special_tokens_map.json
cycy233-misllm-v4-3-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/cycy233-misllm-v4-3-v1/flywheel_model.0.safetensors
cycy233-misllm-v4-3-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
cycy233-misllm-v4-3-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.
cycy233-misllm-v4-3-v1-mkmlizer: warnings.warn(
cycy233-misllm-v4-3-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.
cycy233-misllm-v4-3-v1-mkmlizer: warnings.warn(
cycy233-misllm-v4-3-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()
cycy233-misllm-v4-3-v1-mkmlizer: return self.fget.__get__(instance, owner)()
cycy233-misllm-v4-3-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
cycy233-misllm-v4-3-v1-mkmlizer: Saving duration: 0.394s
cycy233-misllm-v4-3-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.589s
cycy233-misllm-v4-3-v1-mkmlizer: creating bucket guanaco-reward-models
cycy233-misllm-v4-3-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
cycy233-misllm-v4-3-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/cycy233-misllm-v4-3-v1_reward
cycy233-misllm-v4-3-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/cycy233-misllm-v4-3-v1_reward/config.json
cycy233-misllm-v4-3-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/cycy233-misllm-v4-3-v1_reward/tokenizer_config.json
cycy233-misllm-v4-3-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/cycy233-misllm-v4-3-v1_reward/special_tokens_map.json
cycy233-misllm-v4-3-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/cycy233-misllm-v4-3-v1_reward/vocab.json
cycy233-misllm-v4-3-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/cycy233-misllm-v4-3-v1_reward/merges.txt
cycy233-misllm-v4-3-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/cycy233-misllm-v4-3-v1_reward/tokenizer.json
cycy233-misllm-v4-3-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/cycy233-misllm-v4-3-v1_reward/reward.tensors
Job cycy233-misllm-v4-3-v1-mkmlizer completed after 62.89s with status: succeeded
Stopping job with name cycy233-misllm-v4-3-v1-mkmlizer
Pipeline stage MKMLizer completed in 63.23s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service cycy233-misllm-v4-3-v1
Waiting for inference service cycy233-misllm-v4-3-v1 to be ready
Inference service cycy233-misllm-v4-3-v1 ready after 30.179245471954346s
Pipeline stage ISVCDeployer completed in 35.77s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.0877456665039062s
Received healthy response to inference request in 1.129183053970337s
Received healthy response to inference request in 1.053544521331787s
Received healthy response to inference request in 1.1906685829162598s
Received healthy response to inference request in 0.9531652927398682s
5 requests
0 failed requests
5th percentile: 0.973241138458252
10th percentile: 0.9933169841766357
20th percentile: 1.0334686756134033
30th percentile: 1.0686722278594971
40th percentile: 1.098927640914917
50th percentile: 1.129183053970337
60th percentile: 1.153777265548706
70th percentile: 1.1783714771270752
80th percentile: 1.3700839996337892
90th percentile: 1.7289148330688477
95th percentile: 1.9083302497863768
99th percentile: 2.0518625831604003
mean time: 1.2828614234924316
Pipeline stage StressChecker completed in 7.43s
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
cycy233-misllm-v4-3_v1 status is now deployed due to DeploymentManager action
cycy233-misllm-v4-3_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics