developer_uid: Trace2333
submission_id: trace2333-fd-llama3-v4_v4
model_name: trace2333-fd-llama3-v4_v4
model_group: Trace2333/fd_llama3_v4
status: torndown
timestamp: 2024-08-07T06:08:45+00:00
num_battles: 16229
num_wins: 8389
celo_rating: 1225.92
family_friendly_score: 0.0
submission_type: basic
model_repo: Trace2333/fd_llama3_v4
model_architecture: LlamaForCausalLM
reward_repo: Jellywibble/gpt2_xl_pairwise_89m_step_347634
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: trace2333-fd-llama3-v4_v4
is_internal_developer: False
language_model: Trace2333/fd_llama3_v4
model_size: 8B
ranking_group: single
us_pacific_date: 2024-08-06
win_ratio: 0.5169141659991373
generation_params: {'temperature': 1.15, 'top_p': 1.0, 'min_p': 0.06, 'top_k': 100, 'presence_penalty': 0.0, 'frequency_penalty': 0.1, 'stopping_words': ['\n', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64, 'reward_max_token_input': 256}
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: {'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'}
Resubmit model
Running pipeline stage MKMLizer
Starting job with name trace2333-fd-llama3-v4-v4-mkmlizer
Waiting for job on trace2333-fd-llama3-v4-v4-mkmlizer to finish
trace2333-fd-llama3-v4-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
trace2333-fd-llama3-v4-v4-mkmlizer: ║ _____ __ __ ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ /___/ ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ Version: 0.9.9 ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ https://mk1.ai ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ The license key for the current software has been verified as ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ belonging to: ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ Chai Research Corp. ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
trace2333-fd-llama3-v4-v4-mkmlizer: ║ ║
trace2333-fd-llama3-v4-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
trace2333-fd-llama3-v4-v4-mkmlizer: Downloaded to shared memory in 41.320s
trace2333-fd-llama3-v4-v4-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmprl9lnyyd, device:0
trace2333-fd-llama3-v4-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
trace2333-fd-llama3-v4-v4-mkmlizer: quantized model in 28.819s
trace2333-fd-llama3-v4-v4-mkmlizer: Processed model Trace2333/fd_llama3_v4 in 70.139s
trace2333-fd-llama3-v4-v4-mkmlizer: creating bucket guanaco-mkml-models
trace2333-fd-llama3-v4-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
trace2333-fd-llama3-v4-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v4
trace2333-fd-llama3-v4-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v4/special_tokens_map.json
trace2333-fd-llama3-v4-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v4/config.json
trace2333-fd-llama3-v4-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v4/tokenizer_config.json
trace2333-fd-llama3-v4-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v4/tokenizer.json
trace2333-fd-llama3-v4-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v4/flywheel_model.0.safetensors
trace2333-fd-llama3-v4-v4-mkmlizer: loading reward model from Jellywibble/gpt2_xl_pairwise_89m_step_347634
trace2333-fd-llama3-v4-v4-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 5/291 [00:00<00:10, 27.62it/s] Loading 0: 4%|▍ | 12/291 [00:00<00:06, 44.08it/s] Loading 0: 6%|▌ | 17/291 [00:00<00:06, 40.91it/s] Loading 0: 8%|▊ | 22/291 [00:00<00:06, 39.91it/s] Loading 0: 9%|▉ | 27/291 [00:00<00:06, 41.51it/s] Loading 0: 11%|█ | 32/291 [00:00<00:06, 39.75it/s] Loading 0: 13%|█▎ | 37/291 [00:01<00:09, 26.47it/s] Loading 0: 14%|█▍ | 41/291 [00:01<00:09, 26.50it/s] Loading 0: 16%|█▋ | 48/291 [00:01<00:06, 34.76it/s] Loading 0: 18%|█▊ | 53/291 [00:01<00:06, 35.68it/s] Loading 0: 20%|█▉ | 58/291 [00:01<00:06, 36.51it/s] Loading 0: 21%|██▏ | 62/291 [00:01<00:06, 37.30it/s] Loading 0: 23%|██▎ | 66/291 [00:01<00:06, 37.09it/s] Loading 0: 24%|██▍ | 70/291 [00:01<00:06, 35.67it/s] Loading 0: 25%|██▌ | 74/291 [00:02<00:06, 35.37it/s] Loading 0: 27%|██▋ | 78/291 [00:02<00:06, 35.41it/s] Loading 0: 28%|██▊ | 82/291 [00:02<00:08, 24.45it/s] Loading 0: 30%|██▉ | 86/291 [00:02<00:07, 27.52it/s] Loading 0: 31%|███▏ | 91/291 [00:02<00:06, 30.31it/s] Loading 0: 33%|███▎ | 95/291 [00:02<00:06, 32.33it/s] Loading 0: 34%|███▍ | 100/291 [00:02<00:05, 34.11it/s] Loading 0: 36%|███▌ | 104/291 [00:03<00:05, 35.50it/s] Loading 0: 37%|███▋ | 108/291 [00:03<00:05, 36.28it/s] Loading 0: 38%|███▊ | 112/291 [00:03<00:05, 34.41it/s] Loading 0: 40%|███▉ | 116/291 [00:03<00:05, 31.27it/s] Loading 0: 42%|████▏ | 122/291 [00:03<00:04, 36.14it/s] Loading 0: 44%|████▎ | 127/291 [00:03<00:04, 34.08it/s] Loading 0: 46%|████▌ | 133/291 [00:04<00:05, 28.63it/s] Loading 0: 47%|████▋ | 137/291 [00:04<00:05, 29.34it/s] Loading 0: 48%|████▊ | 141/291 [00:04<00:05, 27.81it/s] Loading 0: 51%|█████ | 147/291 [00:04<00:04, 32.96it/s] Loading 0: 52%|█████▏ | 151/291 [00:04<00:04, 32.21it/s] Loading 0: 54%|█████▎ | 156/291 [00:04<00:03, 34.84it/s] Loading 0: 55%|█████▍ | 160/291 [00:04<00:03, 33.53it/s] Loading 0: 57%|█████▋ | 165/291 [00:04<00:03, 35.87it/s] Loading 0: 58%|█████▊ | 169/291 [00:05<00:03, 34.28it/s] Loading 0: 60%|█████▉ | 174/291 [00:05<00:03, 37.01it/s] Loading 0: 61%|██████ | 178/291 [00:05<00:03, 35.29it/s] Loading 0: 63%|██████▎ | 183/291 [00:05<00:02, 38.27it/s] Loading 0: 64%|██████▍ | 187/291 [00:05<00:03, 26.76it/s] Loading 0: 66%|██████▌ | 191/291 [00:05<00:03, 27.37it/s] Loading 0: 67%|██████▋ | 195/291 [00:05<00:03, 26.09it/s] Loading 0: 69%|██████▉ | 201/291 [00:06<00:02, 31.36it/s] Loading 0: 70%|███████ | 205/291 [00:06<00:02, 31.13it/s] Loading 0: 72%|███████▏ | 210/291 [00:06<00:02, 34.13it/s] Loading 0: 74%|███████▎ | 214/291 [00:06<00:02, 33.55it/s] Loading 0: 75%|███████▌ | 219/291 [00:06<00:01, 36.37it/s] Loading 0: 77%|███████▋ | 223/291 [00:06<00:01, 34.64it/s] Loading 0: 78%|███████▊ | 227/291 [00:06<00:01, 34.53it/s] Loading 0: 79%|███████▉ | 231/291 [00:06<00:01, 34.48it/s] Loading 0: 81%|████████ | 235/291 [00:07<00:02, 24.49it/s] Loading 0: 82%|████████▏ | 239/291 [00:07<00:02, 24.90it/s] Loading 0: 85%|████████▍ | 246/291 [00:07<00:01, 32.81it/s] Loading 0: 86%|████████▌ | 250/291 [00:07<00:01, 32.28it/s] Loading 0: 88%|████████▊ | 255/291 [00:07<00:01, 35.32it/s] Loading 0: 89%|████████▉ | 259/291 [00:07<00:00, 34.16it/s] Loading 0: 91%|█████████ | 264/291 [00:08<00:00, 36.53it/s] Loading 0: 92%|█████████▏| 268/291 [00:08<00:00, 34.80it/s] Loading 0: 94%|█████████▍| 273/291 [00:08<00:00, 37.32it/s] Loading 0: 95%|█████████▌| 277/291 [00:08<00:00, 34.48it/s] Loading 0: 97%|█████████▋| 281/291 [00:08<00:00, 33.97it/s] Loading 0: 98%|█████████▊| 286/291 [00:13<00:01, 2.60it/s] Loading 0: 99%|█████████▉| 289/291 [00:14<00:00, 3.25it/s] /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:957: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
trace2333-fd-llama3-v4-v4-mkmlizer: warnings.warn(
trace2333-fd-llama3-v4-v4-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:785: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
trace2333-fd-llama3-v4-v4-mkmlizer: warnings.warn(
trace2333-fd-llama3-v4-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.
trace2333-fd-llama3-v4-v4-mkmlizer: warnings.warn(
trace2333-fd-llama3-v4-v4-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
trace2333-fd-llama3-v4-v4-mkmlizer: Saving duration: 1.387s
trace2333-fd-llama3-v4-v4-mkmlizer: Processed model Jellywibble/gpt2_xl_pairwise_89m_step_347634 in 11.089s
trace2333-fd-llama3-v4-v4-mkmlizer: creating bucket guanaco-reward-models
trace2333-fd-llama3-v4-v4-mkmlizer: Bucket 's3://guanaco-reward-models/' created
trace2333-fd-llama3-v4-v4-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/trace2333-fd-llama3-v4-v4_reward
trace2333-fd-llama3-v4-v4-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v4_reward/config.json
trace2333-fd-llama3-v4-v4-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v4_reward/special_tokens_map.json
trace2333-fd-llama3-v4-v4-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v4_reward/tokenizer_config.json
trace2333-fd-llama3-v4-v4-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/trace2333-fd-llama3-v4-v4_reward/merges.txt
trace2333-fd-llama3-v4-v4-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v4_reward/vocab.json
trace2333-fd-llama3-v4-v4-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v4_reward/tokenizer.json
trace2333-fd-llama3-v4-v4-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/trace2333-fd-llama3-v4-v4_reward/reward.tensors
Job trace2333-fd-llama3-v4-v4-mkmlizer completed after 114.97s with status: succeeded
Stopping job with name trace2333-fd-llama3-v4-v4-mkmlizer
Pipeline stage MKMLizer completed in 116.06s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service trace2333-fd-llama3-v4-v4
Waiting for inference service trace2333-fd-llama3-v4-v4 to be ready
Inference service trace2333-fd-llama3-v4-v4 ready after 191.28001976013184s
Pipeline stage ISVCDeployer completed in 192.86s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3282251358032227s
Received healthy response to inference request in 1.3986718654632568s
Received healthy response to inference request in 1.4292495250701904s
Received healthy response to inference request in 1.4043409824371338s
Received healthy response to inference request in 1.3909587860107422s
5 requests
0 failed requests
5th percentile: 1.3925014019012452
10th percentile: 1.3940440177917481
20th percentile: 1.3971292495727539
30th percentile: 1.3998056888580321
40th percentile: 1.402073335647583
50th percentile: 1.4043409824371338
60th percentile: 1.4143043994903564
70th percentile: 1.4242678165435791
80th percentile: 1.609044647216797
90th percentile: 1.9686348915100098
95th percentile: 2.148430013656616
99th percentile: 2.2922661113739013
mean time: 1.590289258956909
Pipeline stage StressChecker completed in 8.65s
trace2333-fd-llama3-v4_v4 status is now deployed due to DeploymentManager action
trace2333-fd-llama3-v4_v4 status is now inactive due to auto deactivation removed underperforming models
trace2333-fd-llama3-v4_v4 status is now torndown due to DeploymentManager action