submission_id: bbchicago-test-120k-pref-dpo_v4
developer_uid: Bbbrun0
best_of: 16
celo_rating: 1214.6
display_name: bbchicago-test_120k_pref_dpo_v4
family_friendly_score: 0.0
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}
generation_params: {'temperature': 0.95, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|end_header_id|>', '<|eot_id|>', '\n\n{user_name}'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}
is_internal_developer: False
language_model: BBChicago/test_120k_pref_dpo
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: BBChicago/test_120k_pref
model_name: bbchicago-test_120k_pref_dpo_v4
model_num_parameters: 8030261248.0
model_repo: BBChicago/test_120k_pref_dpo
model_size: 8B
num_battles: 17549
num_wins: 9430
ranking_group: single
reward_formatter: {'bot_template': '{bot_name}: {message}\n', 'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'response_template': '{bot_name}:', 'truncate_by_message': False, 'user_template': '{user_name}: {message}\n'}
reward_repo: Jellywibble/gpt2_xl_pairwise_89m_step_347634
status: torndown
submission_type: basic
timestamp: 2024-07-18T01:43:23+00:00
us_pacific_date: 2024-07-17
win_ratio: 0.5373525557011796
Resubmit model
Running pipeline stage MKMLizer
Starting job with name bbchicago-test-120k-pref-dpo-v4-mkmlizer
Waiting for job on bbchicago-test-120k-pref-dpo-v4-mkmlizer to finish
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ _____ __ __ ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ /___/ ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ Version: 0.9.5.post3 ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ https://mk1.ai ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ The license key for the current software has been verified as ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ belonging to: ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ Chai Research Corp. ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ║ ║
bbchicago-test-120k-pref-dpo-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
bbchicago-test-120k-pref-dpo-v4-mkmlizer: Downloaded to shared memory in 20.917s
bbchicago-test-120k-pref-dpo-v4-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpp37fiv1a, device:0
bbchicago-test-120k-pref-dpo-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
bbchicago-test-120k-pref-dpo-v4-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 4%|▍ | 13/291 [00:00<00:02, 127.81it/s] Loading 0: 11%|█ | 31/291 [00:00<00:01, 148.23it/s] Loading 0: 16%|█▌ | 46/291 [00:00<00:01, 148.02it/s] Loading 0: 21%|██ | 61/291 [00:00<00:01, 141.16it/s] Loading 0: 26%|██▌ | 76/291 [00:00<00:01, 143.30it/s] Loading 0: 31%|███▏ | 91/291 [00:00<00:02, 74.31it/s] Loading 0: 36%|███▌ | 104/291 [00:01<00:02, 81.99it/s] Loading 0: 42%|████▏ | 121/291 [00:01<00:01, 97.69it/s] Loading 0: 47%|████▋ | 138/291 [00:01<00:01, 112.86it/s] Loading 0: 53%|█████▎ | 153/291 [00:01<00:01, 121.09it/s] Loading 0: 57%|█████▋ | 167/291 [00:01<00:01, 122.94it/s] Loading 0: 64%|██████▎ | 185/291 [00:01<00:00, 137.41it/s] Loading 0: 69%|██████▊ | 200/291 [00:01<00:01, 80.10it/s] Loading 0: 74%|███████▎ | 214/291 [00:02<00:00, 90.86it/s] Loading 0: 79%|███████▉ | 230/291 [00:02<00:00, 102.87it/s] Loading 0: 85%|████████▌ | 248/291 [00:02<00:00, 116.18it/s] Loading 0: 91%|█████████▏| 266/291 [00:02<00:00, 126.73it/s] Loading 0: 98%|█████████▊| 284/291 [00:02<00:00, 138.79it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
bbchicago-test-120k-pref-dpo-v4-mkmlizer: quantized model in 23.759s
bbchicago-test-120k-pref-dpo-v4-mkmlizer: Processed model BBChicago/test_120k_pref_dpo in 44.677s
bbchicago-test-120k-pref-dpo-v4-mkmlizer: creating bucket guanaco-mkml-models
bbchicago-test-120k-pref-dpo-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
bbchicago-test-120k-pref-dpo-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/bbchicago-test-120k-pref-dpo-v4
bbchicago-test-120k-pref-dpo-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/bbchicago-test-120k-pref-dpo-v4/config.json
bbchicago-test-120k-pref-dpo-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/bbchicago-test-120k-pref-dpo-v4/special_tokens_map.json
bbchicago-test-120k-pref-dpo-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/bbchicago-test-120k-pref-dpo-v4/tokenizer_config.json
bbchicago-test-120k-pref-dpo-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/bbchicago-test-120k-pref-dpo-v4/tokenizer.json
bbchicago-test-120k-pref-dpo-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/bbchicago-test-120k-pref-dpo-v4/flywheel_model.0.safetensors
bbchicago-test-120k-pref-dpo-v4-mkmlizer: loading reward model from Jellywibble/gpt2_xl_pairwise_89m_step_347634
bbchicago-test-120k-pref-dpo-v4-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:950: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
bbchicago-test-120k-pref-dpo-v4-mkmlizer: warnings.warn(
bbchicago-test-120k-pref-dpo-v4-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:778: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
bbchicago-test-120k-pref-dpo-v4-mkmlizer: warnings.warn(
bbchicago-test-120k-pref-dpo-v4-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:469: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
bbchicago-test-120k-pref-dpo-v4-mkmlizer: warnings.warn(
bbchicago-test-120k-pref-dpo-v4-mkmlizer: Saving duration: 1.778s
bbchicago-test-120k-pref-dpo-v4-mkmlizer: Processed model Jellywibble/gpt2_xl_pairwise_89m_step_347634 in 11.336s
bbchicago-test-120k-pref-dpo-v4-mkmlizer: creating bucket guanaco-reward-models
bbchicago-test-120k-pref-dpo-v4-mkmlizer: Bucket 's3://guanaco-reward-models/' created
bbchicago-test-120k-pref-dpo-v4-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/bbchicago-test-120k-pref-dpo-v4_reward
bbchicago-test-120k-pref-dpo-v4-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/bbchicago-test-120k-pref-dpo-v4_reward/special_tokens_map.json
bbchicago-test-120k-pref-dpo-v4-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/bbchicago-test-120k-pref-dpo-v4_reward/config.json
bbchicago-test-120k-pref-dpo-v4-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/bbchicago-test-120k-pref-dpo-v4_reward/tokenizer_config.json
bbchicago-test-120k-pref-dpo-v4-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/bbchicago-test-120k-pref-dpo-v4_reward/merges.txt
bbchicago-test-120k-pref-dpo-v4-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/bbchicago-test-120k-pref-dpo-v4_reward/vocab.json
bbchicago-test-120k-pref-dpo-v4-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/bbchicago-test-120k-pref-dpo-v4_reward/tokenizer.json
bbchicago-test-120k-pref-dpo-v4-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/bbchicago-test-120k-pref-dpo-v4_reward/reward.tensors
Job bbchicago-test-120k-pref-dpo-v4-mkmlizer completed after 105.56s with status: succeeded
Stopping job with name bbchicago-test-120k-pref-dpo-v4-mkmlizer
Pipeline stage MKMLizer completed in 106.87s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service bbchicago-test-120k-pref-dpo-v4
Waiting for inference service bbchicago-test-120k-pref-dpo-v4 to be ready
Failed to get response for submission hastagaras-dirtybu8bl3-r_5954_v3: ('http://hastagaras-dirtybu8bl3-r-5954-v3-predictor-default.tenant-chaiml-guanaco.knative.ord1.coreweave.cloud/v1/models/GPT-J-6B-lit-v2:predict', 'request timeout')
Inference service bbchicago-test-120k-pref-dpo-v4 ready after 162.8691647052765s
Pipeline stage ISVCDeployer completed in 164.50s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.19539737701416s
Received healthy response to inference request in 1.4007635116577148s
Received healthy response to inference request in 1.3881158828735352s
Received healthy response to inference request in 1.394545555114746s
Received healthy response to inference request in 1.411219835281372s
5 requests
0 failed requests
5th percentile: 1.3894018173217773
10th percentile: 1.3906877517700196
20th percentile: 1.393259620666504
30th percentile: 1.3957891464233398
40th percentile: 1.3982763290405273
50th percentile: 1.4007635116577148
60th percentile: 1.4049460411071777
70th percentile: 1.4091285705566405
80th percentile: 1.5680553436279299
90th percentile: 1.881726360321045
95th percentile: 2.0385618686676024
99th percentile: 2.1640302753448486
mean time: 1.5580084323883057
Pipeline stage StressChecker completed in 8.43s
bbchicago-test-120k-pref-dpo_v4 status is now deployed due to DeploymentManager action
bbchicago-test-120k-pref-dpo_v4 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of bbchicago-test-120k-pref-dpo_v4
Running pipeline stage ISVCDeleter
Checking if service bbchicago-test-120k-pref-dpo-v4 is running
Tearing down inference service bbchicago-test-120k-pref-dpo-v4
Service bbchicago-test-120k-pref-dpo-v4 has been torndown
Pipeline stage ISVCDeleter completed in 5.54s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key bbchicago-test-120k-pref-dpo-v4/config.json from bucket guanaco-mkml-models
Deleting key bbchicago-test-120k-pref-dpo-v4/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key bbchicago-test-120k-pref-dpo-v4/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key bbchicago-test-120k-pref-dpo-v4/tokenizer.json from bucket guanaco-mkml-models
Deleting key bbchicago-test-120k-pref-dpo-v4/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key bbchicago-test-120k-pref-dpo-v4_reward/config.json from bucket guanaco-reward-models
Deleting key bbchicago-test-120k-pref-dpo-v4_reward/merges.txt from bucket guanaco-reward-models
Deleting key bbchicago-test-120k-pref-dpo-v4_reward/reward.tensors from bucket guanaco-reward-models
Deleting key bbchicago-test-120k-pref-dpo-v4_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key bbchicago-test-120k-pref-dpo-v4_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key bbchicago-test-120k-pref-dpo-v4_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key bbchicago-test-120k-pref-dpo-v4_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.42s
bbchicago-test-120k-pref-dpo_v4 status is now torndown due to DeploymentManager action