submission_id: cycy233-l3-klee-v2-e3-0_v1
developer_uid: shiroe40
status: inactive
model_repo: cycy233/L3-klee-v2-e3.0
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.0, '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': 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: 2024-07-02T02:23:29+00:00
model_name: auto
model_group: cycy233/L3-klee-v2-e3.0
num_battles: 10872
num_wins: 5565
celo_rating: 1183.94
propriety_score: 0.7152368270876927
propriety_total_count: 5257.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: auto
ineligible_reason: None
language_model: cycy233/L3-klee-v2-e3.0
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-07-01
win_ratio: 0.5118653421633554
Resubmit model
Running pipeline stage MKMLizer
Starting job with name cycy233-l3-klee-v2-e3-0-v1-mkmlizer
Waiting for job on cycy233-l3-klee-v2-e3-0-v1-mkmlizer to finish
Failed to get response for submission pawankrd-cosmosrp_v22: ('http://pawankrd-cosmosrp-v22-predictor-default.tenant-chaiml-guanaco.knative.ord1.coreweave.cloud/v1/models/GPT-J-6B-lit-v2:predict', 'readfrom tcp 127.0.0.1:41448->127.0.0.1:8080: write tcp 127.0.0.1:41448->127.0.0.1:8080: use of closed network connection\n')
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ _____ __ __ ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ /___/ ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ Version: 0.8.14 ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ https://mk1.ai ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ The license key for the current software has been verified as ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ belonging to: ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ Chai Research Corp. ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ║ ║
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: Downloaded to shared memory in 37.585s
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: quantizing model to /dev/shm/model_cache
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Retrying (%r) after connection broken by '%r': %s
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 29%|██▊ | 83/291 [00:01<00:03, 64.88it/s] Loading 0: 64%|██████▍ | 187/291 [00:02<00:01, 78.71it/s] Loading 0: 99%|█████████▊| 287/291 [00:02<00:00, 114.14it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: quantized model in 24.594s
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: Processed model cycy233/L3-klee-v2-e3.0 in 62.179s
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: creating bucket guanaco-mkml-models
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/cycy233-l3-klee-v2-e3-0-v1
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/cycy233-l3-klee-v2-e3-0-v1/config.json
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/cycy233-l3-klee-v2-e3-0-v1/special_tokens_map.json
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/cycy233-l3-klee-v2-e3-0-v1/tokenizer_config.json
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/cycy233-l3-klee-v2-e3-0-v1/tokenizer.json
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/cycy233-l3-klee-v2-e3-0-v1/flywheel_model.0.safetensors
cycy233-l3-klee-v2-e3-0-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.
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: warnings.warn(
cycy233-l3-klee-v2-e3-0-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-l3-klee-v2-e3-0-v1-mkmlizer: return self.fget.__get__(instance, owner)()
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: Saving duration: 0.418s
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.485s
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: creating bucket guanaco-reward-models
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/cycy233-l3-klee-v2-e3-0-v1_reward
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/cycy233-l3-klee-v2-e3-0-v1_reward/config.json
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/cycy233-l3-klee-v2-e3-0-v1_reward/tokenizer_config.json
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/cycy233-l3-klee-v2-e3-0-v1_reward/special_tokens_map.json
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/cycy233-l3-klee-v2-e3-0-v1_reward/merges.txt
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/cycy233-l3-klee-v2-e3-0-v1_reward/vocab.json
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/cycy233-l3-klee-v2-e3-0-v1_reward/tokenizer.json
cycy233-l3-klee-v2-e3-0-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/cycy233-l3-klee-v2-e3-0-v1_reward/reward.tensors
Job cycy233-l3-klee-v2-e3-0-v1-mkmlizer completed after 84.67s with status: succeeded
Stopping job with name cycy233-l3-klee-v2-e3-0-v1-mkmlizer
Pipeline stage MKMLizer completed in 85.63s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.13s
Running pipeline stage ISVCDeployer
Creating inference service cycy233-l3-klee-v2-e3-0-v1
Waiting for inference service cycy233-l3-klee-v2-e3-0-v1 to be ready
Inference service cycy233-l3-klee-v2-e3-0-v1 ready after 40.312907218933105s
Pipeline stage ISVCDeployer completed in 47.80s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.087402105331421s
Received healthy response to inference request in 1.4089102745056152s
Received healthy response to inference request in 1.325617790222168s
Received healthy response to inference request in 1.3131961822509766s
Received healthy response to inference request in 1.438063383102417s
5 requests
0 failed requests
5th percentile: 1.3156805038452148
10th percentile: 1.3181648254394531
20th percentile: 1.3231334686279297
30th percentile: 1.3422762870788574
40th percentile: 1.3755932807922364
50th percentile: 1.4089102745056152
60th percentile: 1.420571517944336
70th percentile: 1.4322327613830566
80th percentile: 1.567931127548218
90th percentile: 1.8276666164398194
95th percentile: 1.95753436088562
99th percentile: 2.0614285564422605
mean time: 1.5146379470825195
Pipeline stage StressChecker completed in 8.80s
cycy233-l3-klee-v2-e3-0_v1 status is now deployed due to DeploymentManager action
cycy233-l3-klee-v2-e3-0_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics