submission_id: trace2333-fd-llama3-v4_v5
developer_uid: Trace2333
best_of: 16
celo_rating: 1189.79
display_name: trace2333-fd-llama3-v4_v5
family_friendly_score: 0.0
formatter: {'memory_template': "<|start_header_id|>system<|end_header_id|>\n\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
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}
is_internal_developer: False
language_model: Trace2333/fd_llama3_v4
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: Trace2333/fd_llama3_v4
model_name: trace2333-fd-llama3-v4_v5
model_num_parameters: 8030261248.0
model_repo: Trace2333/fd_llama3_v4
model_size: 8B
num_battles: 16780
num_wins: 7724
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: 2024-08-07T12:29:53+00:00
us_pacific_date: 2024-08-07
win_ratio: 0.4603098927294398
Resubmit model
Running pipeline stage MKMLizer
Starting job with name trace2333-fd-llama3-v4-v5-mkmlizer
Waiting for job on trace2333-fd-llama3-v4-v5-mkmlizer to finish
Stopping job with name trace2333-fd-llama3-v4-v5-mkmlizer
%s, retrying in %s seconds...
Starting job with name trace2333-fd-llama3-v4-v5-mkmlizer
Waiting for job on trace2333-fd-llama3-v4-v5-mkmlizer to finish
trace2333-fd-llama3-v4-v5-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
trace2333-fd-llama3-v4-v5-mkmlizer: ║ _____ __ __ ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ /___/ ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ Version: 0.9.9 ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ https://mk1.ai ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ The license key for the current software has been verified as ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ belonging to: ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ Chai Research Corp. ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
trace2333-fd-llama3-v4-v5-mkmlizer: ║ ║
trace2333-fd-llama3-v4-v5-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
trace2333-fd-llama3-v4-v5-mkmlizer: Downloaded to shared memory in 44.768s
trace2333-fd-llama3-v4-v5-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmptfyol55n, device:0
trace2333-fd-llama3-v4-v5-mkmlizer: Saving flywheel model at /dev/shm/model_cache
trace2333-fd-llama3-v4-v5-mkmlizer: quantized model in 30.438s
trace2333-fd-llama3-v4-v5-mkmlizer: Processed model Trace2333/fd_llama3_v4 in 75.206s
trace2333-fd-llama3-v4-v5-mkmlizer: creating bucket guanaco-mkml-models
trace2333-fd-llama3-v4-v5-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
trace2333-fd-llama3-v4-v5-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v5
trace2333-fd-llama3-v4-v5-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v5/config.json
trace2333-fd-llama3-v4-v5-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v5/special_tokens_map.json
trace2333-fd-llama3-v4-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v5/tokenizer_config.json
trace2333-fd-llama3-v4-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v5/tokenizer.json
trace2333-fd-llama3-v4-v5-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/trace2333-fd-llama3-v4-v5/flywheel_model.0.safetensors
trace2333-fd-llama3-v4-v5-mkmlizer: loading reward model from Jellywibble/gpt2_xl_pairwise_89m_step_347634
trace2333-fd-llama3-v4-v5-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 5/291 [00:00<00:10, 26.50it/s] Loading 0: 4%|▍ | 12/291 [00:00<00:06, 41.32it/s] Loading 0: 6%|▌ | 17/291 [00:00<00:07, 38.64it/s] Loading 0: 8%|▊ | 22/291 [00:00<00:07, 36.57it/s] Loading 0: 9%|▉ | 26/291 [00:00<00:07, 36.98it/s] Loading 0: 11%|█ | 32/291 [00:00<00:06, 38.21it/s] Loading 0: 12%|█▏ | 36/291 [00:01<00:10, 23.99it/s] Loading 0: 14%|█▍ | 41/291 [00:01<00:09, 25.79it/s] Loading 0: 16%|█▋ | 48/291 [00:01<00:07, 33.45it/s] Loading 0: 18%|█▊ | 52/291 [00:01<00:07, 32.38it/s] Loading 0: 20%|█▉ | 57/291 [00:01<00:06, 34.90it/s] Loading 0: 21%|██ | 61/291 [00:01<00:06, 33.48it/s] Loading 0: 23%|██▎ | 66/291 [00:01<00:06, 35.99it/s] Loading 0: 24%|██▍ | 70/291 [00:02<00:06, 34.13it/s] Loading 0: 25%|██▌ | 74/291 [00:02<00:06, 33.98it/s] Loading 0: 27%|██▋ | 78/291 [00:02<00:06, 33.19it/s] Loading 0: 28%|██▊ | 82/291 [00:02<00:09, 22.11it/s] Loading 0: 30%|██▉ | 86/291 [00:02<00:08, 24.95it/s] Loading 0: 31%|███ | 90/291 [00:02<00:07, 27.41it/s] Loading 0: 32%|███▏ | 94/291 [00:03<00:07, 27.96it/s] Loading 0: 34%|███▍ | 99/291 [00:03<00:06, 31.26it/s] Loading 0: 35%|███▌ | 103/291 [00:03<00:06, 30.17it/s] Loading 0: 37%|███▋ | 108/291 [00:03<00:05, 33.14it/s] Loading 0: 38%|███▊ | 112/291 [00:03<00:05, 31.39it/s] Loading 0: 40%|███▉ | 116/291 [00:03<00:05, 31.10it/s] Loading 0: 42%|████▏ | 122/291 [00:03<00:04, 35.69it/s] Loading 0: 44%|████▎ | 127/291 [00:04<00:04, 33.16it/s] Loading 0: 46%|████▌ | 133/291 [00:04<00:05, 27.50it/s] Loading 0: 47%|████▋ | 137/291 [00:04<00:05, 27.92it/s] Loading 0: 48%|████▊ | 140/291 [00:04<00:06, 24.56it/s] Loading 0: 50%|████▉ | 145/291 [00:04<00:04, 29.30it/s] Loading 0: 51%|█████ | 149/291 [00:04<00:05, 26.67it/s] Loading 0: 53%|█████▎ | 154/291 [00:05<00:04, 31.12it/s] Loading 0: 54%|█████▍ | 158/291 [00:05<00:04, 27.86it/s] Loading 0: 56%|█████▌ | 163/291 [00:05<00:03, 32.59it/s] Loading 0: 57%|█████▋ | 167/291 [00:05<00:04, 28.32it/s] Loading 0: 59%|█████▉ | 172/291 [00:05<00:03, 32.60it/s] Loading 0: 60%|██████ | 176/291 [00:05<00:04, 28.25it/s] Loading 0: 62%|██████▏ | 181/291 [00:05<00:03, 31.36it/s] Loading 0: 64%|██████▎ | 185/291 [00:06<00:03, 31.91it/s] Loading 0: 65%|██████▍ | 189/291 [00:06<00:05, 19.32it/s] Loading 0: 67%|██████▋ | 194/291 [00:06<00:04, 21.41it/s] Loading 0: 68%|██████▊ | 199/291 [00:06<00:03, 25.99it/s] Loading 0: 70%|██████▉ | 203/291 [00:06<00:03, 24.35it/s] Loading 0: 71%|███████▏ | 208/291 [00:07<00:03, 27.66it/s] Loading 0: 73%|███████▎ | 212/291 [00:07<00:03, 25.72it/s] Loading 0: 75%|███████▍ | 217/291 [00:07<00:02, 30.48it/s] Loading 0: 76%|███████▌ | 221/291 [00:07<00:02, 27.60it/s] Loading 0: 78%|███████▊ | 226/291 [00:07<00:02, 31.85it/s] Loading 0: 79%|███████▉ | 230/291 [00:07<00:02, 28.91it/s] Loading 0: 80%|████████ | 234/291 [00:08<00:02, 21.35it/s] Loading 0: 82%|████████▏ | 239/291 [00:08<00:02, 23.22it/s] Loading 0: 84%|████████▍ | 244/291 [00:08<00:01, 27.04it/s] Loading 0: 85%|████████▌ | 248/291 [00:08<00:01, 25.37it/s] Loading 0: 88%|████████▊ | 255/291 [00:08<00:01, 31.91it/s] Loading 0: 89%|████████▉ | 259/291 [00:08<00:01, 30.59it/s] Loading 0: 91%|█████████ | 264/291 [00:09<00:00, 32.59it/s] Loading 0: 92%|█████████▏| 268/291 [00:09<00:00, 31.27it/s] Loading 0: 94%|█████████▍| 273/291 [00:09<00:00, 33.35it/s] Loading 0: 95%|█████████▌| 277/291 [00:09<00:00, 30.74it/s] Loading 0: 97%|█████████▋| 281/291 [00:09<00:00, 30.82it/s] Loading 0: 98%|█████████▊| 286/291 [00:15<00:01, 2.54it/s] Loading 0: 99%|█████████▉| 289/291 [00:15<00:00, 3.16it/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-v5-mkmlizer: warnings.warn(
trace2333-fd-llama3-v4-v5-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-v5-mkmlizer: warnings.warn(
trace2333-fd-llama3-v4-v5-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-v5-mkmlizer: warnings.warn(
trace2333-fd-llama3-v4-v5-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
trace2333-fd-llama3-v4-v5-mkmlizer: Saving duration: 1.425s
trace2333-fd-llama3-v4-v5-mkmlizer: Processed model Jellywibble/gpt2_xl_pairwise_89m_step_347634 in 12.499s
trace2333-fd-llama3-v4-v5-mkmlizer: creating bucket guanaco-reward-models
trace2333-fd-llama3-v4-v5-mkmlizer: Bucket 's3://guanaco-reward-models/' created
trace2333-fd-llama3-v4-v5-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/trace2333-fd-llama3-v4-v5_reward
trace2333-fd-llama3-v4-v5-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v5_reward/config.json
trace2333-fd-llama3-v4-v5-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v5_reward/special_tokens_map.json
trace2333-fd-llama3-v4-v5-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v5_reward/tokenizer_config.json
trace2333-fd-llama3-v4-v5-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/trace2333-fd-llama3-v4-v5_reward/merges.txt
trace2333-fd-llama3-v4-v5-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v5_reward/vocab.json
trace2333-fd-llama3-v4-v5-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/trace2333-fd-llama3-v4-v5_reward/tokenizer.json
Job trace2333-fd-llama3-v4-v5-mkmlizer completed after 125.4s with status: succeeded
Stopping job with name trace2333-fd-llama3-v4-v5-mkmlizer
Pipeline stage MKMLizer completed in 127.02s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service trace2333-fd-llama3-v4-v5
Waiting for inference service trace2333-fd-llama3-v4-v5 to be ready
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Inference service trace2333-fd-llama3-v4-v5 ready after 181.07218265533447s
Pipeline stage ISVCDeployer completed in 182.64s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.342233657836914s
Received healthy response to inference request in 1.3925564289093018s
Received healthy response to inference request in 1.384955644607544s
Received healthy response to inference request in 1.3954682350158691s
Received healthy response to inference request in 1.3868565559387207s
5 requests
0 failed requests
5th percentile: 1.3853358268737792
10th percentile: 1.3857160091400147
20th percentile: 1.3864763736724854
30th percentile: 1.387996530532837
40th percentile: 1.3902764797210694
50th percentile: 1.3925564289093018
60th percentile: 1.3937211513519288
70th percentile: 1.3948858737945558
80th percentile: 1.5848213195800782
90th percentile: 1.9635274887084961
95th percentile: 2.152880573272705
99th percentile: 2.304363040924072
mean time: 1.58041410446167
Pipeline stage StressChecker completed in 8.55s
trace2333-fd-llama3-v4_v5 status is now deployed due to DeploymentManager action
trace2333-fd-llama3-v4_v5 status is now inactive due to auto deactivation removed underperforming models
trace2333-fd-llama3-v4_v5 status is now torndown due to DeploymentManager action