trace2333-andrij-llama2-v1

developer_uid: Trace2333

submission_id: trace2333-andrij-llama2-v1_v1

model_name: trace2333-andrij-llama2-v1_v1

model_group: Trace2333/andrij_llama2_

status: torndown

timestamp: 2024-07-18T02:22:56+00:00

num_battles: 16876

num_wins: 7140

celo_rating: 1132.42

family_friendly_score: 0.0

submission_type: basic

model_repo: Trace2333/andrij_llama2_v1

model_architecture: LlamaForCausalLM

reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2

model_num_parameters: 6738415616.0

best_of: 16

max_input_tokens: 512

max_output_tokens: 64

display_name: trace2333-andrij-llama2-v1_v1

is_internal_developer: False

language_model: Trace2333/andrij_llama2_v1

model_size: 7B

ranking_group: single

us_pacific_date: 2024-07-17

win_ratio: 0.42308603934581657

generation_params: {'temperature': 1.05, 'top_p': 1.0, 'min_p': 0.06, 'top_k': 40, 'presence_penalty': 0.2, 'frequency_penalty': 0.1, 'stopping_words': ['</s>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}

formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': 'You: {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-andrij-llama2-v1-v1-mkmlizer
Waiting for job on trace2333-andrij-llama2-v1-v1-mkmlizer to finish
trace2333-andrij-llama2-v1-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
trace2333-andrij-llama2-v1-v1-mkmlizer: ║     _____            __           __                                ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║       /___/                                                         ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║                                                                     ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║  Version: 0.9.5.post3                                               ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║  https://mk1.ai                                                     ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║                                                                     ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║  belonging to:                                                      ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║                                                                     ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║  Chai Research Corp.                                                ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ║                                                                     ║
trace2333-andrij-llama2-v1-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
trace2333-andrij-llama2-v1-v1-mkmlizer: Downloaded to shared memory in 63.938s
trace2333-andrij-llama2-v1-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp0y94n2ml, device:0
trace2333-andrij-llama2-v1-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
trace2333-andrij-llama2-v1-v1-mkmlizer: quantized model in 21.608s
trace2333-andrij-llama2-v1-v1-mkmlizer: Processed model Trace2333/andrij_llama2_v1 in 85.545s
trace2333-andrij-llama2-v1-v1-mkmlizer: creating bucket guanaco-mkml-models
trace2333-andrij-llama2-v1-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
trace2333-andrij-llama2-v1-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/trace2333-andrij-llama2-v1-v1
trace2333-andrij-llama2-v1-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/trace2333-andrij-llama2-v1-v1/special_tokens_map.json
trace2333-andrij-llama2-v1-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/trace2333-andrij-llama2-v1-v1/config.json
trace2333-andrij-llama2-v1-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/trace2333-andrij-llama2-v1-v1/tokenizer_config.json
trace2333-andrij-llama2-v1-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/trace2333-andrij-llama2-v1-v1/tokenizer.model
trace2333-andrij-llama2-v1-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/trace2333-andrij-llama2-v1-v1/tokenizer.json
trace2333-andrij-llama2-v1-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
trace2333-andrij-llama2-v1-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:04, 63.11it/s]
Loading 0:   5%|▍         | 14/291 [00:00<00:04, 56.38it/s]
Loading 0:   8%|▊         | 22/291 [00:00<00:04, 64.66it/s]
Loading 0:  10%|█         | 30/291 [00:00<00:03, 69.58it/s]
Loading 0:  13%|█▎        | 38/291 [00:00<00:03, 72.57it/s]
Loading 0:  16%|█▌        | 46/291 [00:00<00:03, 67.55it/s]
Loading 0:  18%|█▊        | 53/291 [00:01<00:08, 28.62it/s]
Loading 0:  20%|██        | 59/291 [00:01<00:07, 32.99it/s]
Loading 0:  23%|██▎       | 66/291 [00:01<00:05, 38.18it/s]
Loading 0:  25%|██▍       | 72/291 [00:01<00:05, 41.19it/s]
Loading 0:  27%|██▋       | 80/291 [00:01<00:04, 49.18it/s]
Loading 0:  30%|███       | 88/291 [00:01<00:03, 55.94it/s]
Loading 0:  33%|███▎      | 95/291 [00:01<00:03, 58.95it/s]
Loading 0:  35%|███▌      | 102/291 [00:02<00:03, 57.76it/s]
Loading 0:  37%|███▋      | 109/291 [00:02<00:06, 27.81it/s]
Loading 0:  39%|███▉      | 114/291 [00:02<00:05, 30.54it/s]
Loading 0:  42%|████▏     | 121/291 [00:02<00:04, 34.71it/s]
Loading 0:  44%|████▍     | 129/291 [00:02<00:03, 42.27it/s]
Loading 0:  47%|████▋     | 136/291 [00:03<00:03, 47.53it/s]
Loading 0:  49%|████▉     | 143/291 [00:03<00:02, 51.48it/s]
Loading 0:  52%|█████▏    | 150/291 [00:03<00:02, 54.68it/s]
Loading 0:  55%|█████▍    | 159/291 [00:03<00:04, 32.03it/s]
Loading 0:  57%|█████▋    | 165/291 [00:03<00:03, 35.25it/s]
Loading 0:  59%|█████▉    | 172/291 [00:03<00:02, 41.06it/s]
Loading 0:  62%|██████▏   | 179/291 [00:04<00:02, 45.86it/s]
Loading 0:  64%|██████▎   | 185/291 [00:04<00:02, 48.10it/s]
Loading 0:  66%|██████▋   | 193/291 [00:04<00:01, 55.55it/s]
Loading 0:  69%|██████▉   | 201/291 [00:04<00:01, 61.12it/s]
Loading 0:  71%|███████▏  | 208/291 [00:04<00:01, 62.79it/s]
Loading 0:  74%|███████▍  | 215/291 [00:05<00:02, 31.47it/s]
Loading 0:  76%|███████▌  | 221/291 [00:05<00:01, 35.11it/s]
Loading 0:  79%|███████▊  | 229/291 [00:05<00:01, 43.19it/s]
Loading 0:  81%|████████▏ | 237/291 [00:05<00:01, 50.64it/s]
Loading 0:  84%|████████▍ | 245/291 [00:05<00:00, 57.27it/s]
Loading 0:  87%|████████▋ | 253/291 [00:05<00:00, 58.12it/s]
Loading 0:  89%|████████▉ | 260/291 [00:05<00:00, 59.18it/s]
Loading 0:  92%|█████████▏| 267/291 [00:07<00:02,  9.93it/s]
Loading 0:  94%|█████████▍| 274/291 [00:07<00:01, 13.09it/s]
Loading 0:  97%|█████████▋| 281/291 [00:08<00:00, 17.11it/s]
Loading 0:  99%|█████████▉| 288/291 [00:08<00:00, 21.87it/s]
                                                            
/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.
trace2333-andrij-llama2-v1-v1-mkmlizer:   warnings.warn(
trace2333-andrij-llama2-v1-v1-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.
trace2333-andrij-llama2-v1-v1-mkmlizer:   warnings.warn(
trace2333-andrij-llama2-v1-v1-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-andrij-llama2-v1-v1-mkmlizer:   warnings.warn(
trace2333-andrij-llama2-v1-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
trace2333-andrij-llama2-v1-v1-mkmlizer: Saving duration: 0.524s
trace2333-andrij-llama2-v1-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 8.613s
trace2333-andrij-llama2-v1-v1-mkmlizer: creating bucket guanaco-reward-models
trace2333-andrij-llama2-v1-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
trace2333-andrij-llama2-v1-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/trace2333-andrij-llama2-v1-v1_reward
trace2333-andrij-llama2-v1-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/trace2333-andrij-llama2-v1-v1_reward/config.json
trace2333-andrij-llama2-v1-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/trace2333-andrij-llama2-v1-v1_reward/tokenizer_config.json
trace2333-andrij-llama2-v1-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/trace2333-andrij-llama2-v1-v1_reward/special_tokens_map.json
trace2333-andrij-llama2-v1-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/trace2333-andrij-llama2-v1-v1_reward/merges.txt
trace2333-andrij-llama2-v1-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/trace2333-andrij-llama2-v1-v1_reward/vocab.json
trace2333-andrij-llama2-v1-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/trace2333-andrij-llama2-v1-v1_reward/tokenizer.json
trace2333-andrij-llama2-v1-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/trace2333-andrij-llama2-v1-v1_reward/reward.tensors
Job trace2333-andrij-llama2-v1-v1-mkmlizer completed after 127.53s with status: succeeded
Stopping job with name trace2333-andrij-llama2-v1-v1-mkmlizer
Pipeline stage MKMLizer completed in 128.87s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.12s
Running pipeline stage ISVCDeployer
Creating inference service trace2333-andrij-llama2-v1-v1
Waiting for inference service trace2333-andrij-llama2-v1-v1 to be ready
Inference service trace2333-andrij-llama2-v1-v1 ready after 40.7458598613739s
Pipeline stage ISVCDeployer completed in 43.05s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3094687461853027s
Received healthy response to inference request in 1.1846768856048584s
Received healthy response to inference request in 1.1510505676269531s
Received healthy response to inference request in 1.1894772052764893s
Received healthy response to inference request in 1.1771471500396729s
5 requests
0 failed requests
5th percentile: 1.1562698841094972
10th percentile: 1.161489200592041
20th percentile: 1.1719278335571288
30th percentile: 1.17865309715271
40th percentile: 1.1816649913787842
50th percentile: 1.1846768856048584
60th percentile: 1.1865970134735107
70th percentile: 1.1885171413421631
80th percentile: 1.4134755134582522
90th percentile: 1.8614721298217773
95th percentile: 2.08547043800354
99th percentile: 2.26466908454895
mean time: 1.4023641109466554
Pipeline stage StressChecker completed in 7.73s
trace2333-andrij-llama2-v1_v1 status is now deployed due to DeploymentManager action
trace2333-andrij-llama2-v1_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of trace2333-andrij-llama2-v1_v1
Running pipeline stage ISVCDeleter
Checking if service trace2333-andrij-llama2-v1-v1 is running
Tearing down inference service trace2333-andrij-llama2-v1-v1
Service trace2333-andrij-llama2-v1-v1 has been torndown
Pipeline stage ISVCDeleter completed in 4.39s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key trace2333-andrij-llama2-v1-v1/config.json from bucket guanaco-mkml-models
Deleting key trace2333-andrij-llama2-v1-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key trace2333-andrij-llama2-v1-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key trace2333-andrij-llama2-v1-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key trace2333-andrij-llama2-v1-v1/tokenizer.model from bucket guanaco-mkml-models
Deleting key trace2333-andrij-llama2-v1-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key trace2333-andrij-llama2-v1-v1_reward/config.json from bucket guanaco-reward-models
Deleting key trace2333-andrij-llama2-v1-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key trace2333-andrij-llama2-v1-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key trace2333-andrij-llama2-v1-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key trace2333-andrij-llama2-v1-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key trace2333-andrij-llama2-v1-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key trace2333-andrij-llama2-v1-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.99s
trace2333-andrij-llama2-v1_v1 status is now torndown due to DeploymentManager action