submission_id: chaiml-elo-alignment-run-3_v4
developer_uid: zonemercy
status: inactive
model_repo: ChaiML/elo_alignment_run_3
reward_repo: ChaiML/gpt2_xl_pairwise_89m_step_347634
generation_params: {'temperature': 0.9, 'top_p': 1.0, 'min_p': 0.05, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|eot_id|>'], '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: 2024-07-13T04:54:02+00:00
model_name: chaiml-elo-alignment-run-3_v3
model_group: ChaiML/elo_alignment_run
num_battles: 50315
num_wins: 28626
celo_rating: 1268.38
alignment_score: None
alignment_samples: 0
propriety_score: 0.7322863403944485
propriety_total_count: 5476.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: chaiml-elo-alignment-run-3_v3
ineligible_reason: None
language_model: ChaiML/elo_alignment_run_3
model_size: 8B
reward_model: ChaiML/gpt2_xl_pairwise_89m_step_347634
us_pacific_date: 2024-07-12
win_ratio: 0.5689357050581337
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name chaiml-elo-alignment-run-3-v4-mkmlizer
Waiting for job on chaiml-elo-alignment-run-3-v4-mkmlizer to finish
chaiml-elo-alignment-run-3-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ _____ __ __ ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ /___/ ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ Version: 0.8.14 ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ https://mk1.ai ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ The license key for the current software has been verified as ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ belonging to: ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ Chai Research Corp. ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ║ ║
chaiml-elo-alignment-run-3-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-elo-alignment-run-3-v4-mkmlizer: Downloaded to shared memory in 48.176s
chaiml-elo-alignment-run-3-v4-mkmlizer: quantizing model to /dev/shm/model_cache
chaiml-elo-alignment-run-3-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-elo-alignment-run-3-v4-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 5/291 [00:00<00:06, 44.60it/s] Loading 0: 4%|▍ | 13/291 [00:00<00:04, 64.40it/s] Loading 0: 8%|▊ | 22/291 [00:00<00:04, 65.47it/s] Loading 0: 11%|█▏ | 33/291 [00:00<00:07, 35.08it/s] Loading 0: 13%|█▎ | 38/291 [00:00<00:06, 37.18it/s] Loading 0: 15%|█▍ | 43/291 [00:01<00:06, 38.87it/s] Loading 0: 17%|█▋ | 50/291 [00:01<00:05, 42.78it/s] Loading 0: 20%|██ | 59/291 [00:01<00:04, 49.54it/s] Loading 0: 23%|██▎ | 68/291 [00:01<00:04, 54.53it/s] Loading 0: 26%|██▌ | 76/291 [00:01<00:03, 60.18it/s] Loading 0: 29%|██▊ | 83/291 [00:02<00:06, 31.98it/s] Loading 0: 32%|███▏ | 92/291 [00:02<00:05, 38.95it/s] Loading 0: 35%|███▍ | 101/291 [00:02<00:04, 45.08it/s] Loading 0: 37%|███▋ | 109/291 [00:02<00:03, 51.54it/s] Loading 0: 40%|████ | 117/291 [00:02<00:03, 57.24it/s] Loading 0: 43%|████▎ | 126/291 [00:02<00:02, 60.55it/s] Loading 0: 46%|████▌ | 133/291 [00:03<00:04, 36.18it/s] Loading 0: 48%|████▊ | 139/291 [00:03<00:04, 36.63it/s] Loading 0: 51%|█████ | 147/291 [00:03<00:03, 43.16it/s] Loading 0: 53%|█████▎ | 153/291 [00:03<00:03, 45.38it/s] Loading 0: 55%|█████▍ | 159/291 [00:03<00:03, 43.76it/s] Loading 0: 57%|█████▋ | 166/291 [00:03<00:02, 46.56it/s] Loading 0: 60%|█████▉ | 174/291 [00:03<00:02, 51.31it/s] Loading 0: 62%|██████▏ | 180/291 [00:03<00:02, 51.25it/s] Loading 0: 64%|██████▍ | 186/291 [00:04<00:03, 30.13it/s] Loading 0: 66%|██████▌ | 191/291 [00:04<00:03, 32.86it/s] Loading 0: 67%|██████▋ | 196/291 [00:04<00:02, 35.09it/s] Loading 0: 70%|██████▉ | 203/291 [00:04<00:02, 39.93it/s] Loading 0: 73%|███████▎ | 212/291 [00:04<00:01, 47.57it/s] Loading 0: 76%|███████▌ | 221/291 [00:04<00:01, 53.10it/s] Loading 0: 79%|███████▊ | 229/291 [00:05<00:01, 58.93it/s] Loading 0: 81%|████████ | 236/291 [00:05<00:01, 35.58it/s] Loading 0: 83%|████████▎ | 241/291 [00:05<00:01, 37.06it/s] Loading 0: 85%|████████▌ | 248/291 [00:05<00:01, 41.13it/s] Loading 0: 88%|████████▊ | 257/291 [00:05<00:00, 47.50it/s] Loading 0: 91%|█████████▏| 266/291 [00:05<00:00, 52.81it/s] Loading 0: 95%|█████████▍| 275/291 [00:06<00:00, 56.87it/s] Loading 0: 98%|█████████▊| 286/291 [00:13<00:01, 4.12it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
chaiml-elo-alignment-run-3-v4-mkmlizer: quantized model in 31.870s
chaiml-elo-alignment-run-3-v4-mkmlizer: Processed model ChaiML/elo_alignment_run_3 in 80.047s
chaiml-elo-alignment-run-3-v4-mkmlizer: creating bucket guanaco-mkml-models
chaiml-elo-alignment-run-3-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-elo-alignment-run-3-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-elo-alignment-run-3-v4
chaiml-elo-alignment-run-3-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-elo-alignment-run-3-v4/config.json
chaiml-elo-alignment-run-3-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-elo-alignment-run-3-v4/special_tokens_map.json
chaiml-elo-alignment-run-3-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-elo-alignment-run-3-v4/tokenizer_config.json
chaiml-elo-alignment-run-3-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-elo-alignment-run-3-v4/tokenizer.json
chaiml-elo-alignment-run-3-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-elo-alignment-run-3-v4/flywheel_model.0.safetensors
chaiml-elo-alignment-run-3-v4-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
chaiml-elo-alignment-run-3-v4-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:919: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
chaiml-elo-alignment-run-3-v4-mkmlizer: warnings.warn(
chaiml-elo-alignment-run-3-v4-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
chaiml-elo-alignment-run-3-v4-mkmlizer: warnings.warn(
chaiml-elo-alignment-run-3-v4-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:769: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
chaiml-elo-alignment-run-3-v4-mkmlizer: warnings.warn(
chaiml-elo-alignment-run-3-v4-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-elo-alignment-run-3-v4-mkmlizer: warnings.warn(
chaiml-elo-alignment-run-3-v4-mkmlizer: Downloading shards: 0%| | 0/2 [00:00<?, ?it/s] Downloading shards: 50%|█████ | 1/2 [00:09<00:09, 9.29s/it] Downloading shards: 100%|██████████| 2/2 [00:11<00:00, 5.13s/it] Downloading shards: 100%|██████████| 2/2 [00:11<00:00, 5.75s/it]
chaiml-elo-alignment-run-3-v4-mkmlizer: Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 1.66it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 2.70it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 2.47it/s]
chaiml-elo-alignment-run-3-v4-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
chaiml-elo-alignment-run-3-v4-mkmlizer: Saving duration: 2.188s
chaiml-elo-alignment-run-3-v4-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 16.065s
chaiml-elo-alignment-run-3-v4-mkmlizer: creating bucket guanaco-reward-models
chaiml-elo-alignment-run-3-v4-mkmlizer: Bucket 's3://guanaco-reward-models/' created
chaiml-elo-alignment-run-3-v4-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v4_reward
chaiml-elo-alignment-run-3-v4-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v4_reward/config.json
chaiml-elo-alignment-run-3-v4-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v4_reward/special_tokens_map.json
chaiml-elo-alignment-run-3-v4-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v4_reward/vocab.json
chaiml-elo-alignment-run-3-v4-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v4_reward/merges.txt
chaiml-elo-alignment-run-3-v4-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v4_reward/tokenizer_config.json
chaiml-elo-alignment-run-3-v4-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v4_reward/tokenizer.json
chaiml-elo-alignment-run-3-v4-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/chaiml-elo-alignment-run-3-v4_reward/reward.tensors
Job chaiml-elo-alignment-run-3-v4-mkmlizer completed after 124.63s with status: succeeded
Stopping job with name chaiml-elo-alignment-run-3-v4-mkmlizer
Pipeline stage MKMLizer completed in 125.50s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service chaiml-elo-alignment-run-3-v4
Waiting for inference service chaiml-elo-alignment-run-3-v4 to be ready
Inference service chaiml-elo-alignment-run-3-v4 ready after 40.16545343399048s
Pipeline stage ISVCDeployer completed in 47.33s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.341416120529175s
Received healthy response to inference request in 1.5793178081512451s
Received healthy response to inference request in 1.5426545143127441s
Received healthy response to inference request in 1.5086729526519775s
Received healthy response to inference request in 1.5817699432373047s
5 requests
0 failed requests
5th percentile: 1.5154692649841308
10th percentile: 1.522265577316284
20th percentile: 1.5358582019805909
30th percentile: 1.5499871730804444
40th percentile: 1.5646524906158448
50th percentile: 1.5793178081512451
60th percentile: 1.580298662185669
70th percentile: 1.5812795162200928
80th percentile: 1.7336991786956788
90th percentile: 2.0375576496124266
95th percentile: 2.1894868850708007
99th percentile: 2.3110302734375
mean time: 1.7107662677764892
Pipeline stage StressChecker completed in 9.21s
chaiml-elo-alignment-run-3_v4 status is now deployed due to DeploymentManager action
chaiml-elo-alignment-run-3_v4 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics