jic062-instruct-fail

developer_uid: chace9580

submission_id: jic062-instruct-fail_v2

model_name: jic062-instruct-fail_v2

model_group: jic062/instruct_fail

status: torndown

timestamp: 2024-08-07T02:41:04+00:00

num_battles: 12589

num_wins: 6382

celo_rating: 1220.22

family_friendly_score: 0.0

submission_type: basic

model_repo: jic062/instruct_fail

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: jic062-instruct-fail_v2

is_internal_developer: False

language_model: jic062/instruct_fail

model_size: 8B

ranking_group: single

us_pacific_date: 2024-08-06

win_ratio: 0.5069505123520534

generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|end_of_text|>', '|eot_id|'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64, 'reward_max_token_input': 256}

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: {'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 jic062-instruct-fail-v2-mkmlizer
Waiting for job on jic062-instruct-fail-v2-mkmlizer to finish
jic062-instruct-fail-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-instruct-fail-v2-mkmlizer: ║     _____            __           __                                ║
jic062-instruct-fail-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-instruct-fail-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-instruct-fail-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-instruct-fail-v2-mkmlizer: ║       /___/                                                         ║
jic062-instruct-fail-v2-mkmlizer: ║                                                                     ║
jic062-instruct-fail-v2-mkmlizer: ║  Version: 0.9.9                                                     ║
jic062-instruct-fail-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-instruct-fail-v2-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-instruct-fail-v2-mkmlizer: ║                                                                     ║
jic062-instruct-fail-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-instruct-fail-v2-mkmlizer: ║  belonging to:                                                      ║
jic062-instruct-fail-v2-mkmlizer: ║                                                                     ║
jic062-instruct-fail-v2-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-instruct-fail-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-instruct-fail-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-instruct-fail-v2-mkmlizer: ║                                                                     ║
jic062-instruct-fail-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-instruct-fail-v2-mkmlizer: Downloaded to shared memory in 22.045s
jic062-instruct-fail-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp7iiyhlkp, device:0
jic062-instruct-fail-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-instruct-fail-v2-mkmlizer: quantized model in 26.651s
jic062-instruct-fail-v2-mkmlizer: Processed model jic062/instruct_fail in 48.697s
jic062-instruct-fail-v2-mkmlizer: creating bucket guanaco-mkml-models
jic062-instruct-fail-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-instruct-fail-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-instruct-fail-v2
jic062-instruct-fail-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-instruct-fail-v2/config.json
jic062-instruct-fail-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-instruct-fail-v2/special_tokens_map.json
jic062-instruct-fail-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-instruct-fail-v2/tokenizer_config.json
jic062-instruct-fail-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-instruct-fail-v2/tokenizer.json
jic062-instruct-fail-v2-mkmlizer: loading reward model from Jellywibble/gpt2_xl_pairwise_89m_step_347634
jic062-instruct-fail-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 50.26it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:04, 59.86it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:03, 68.67it/s]
Loading 0:  12%|█▏        | 35/291 [00:00<00:03, 79.15it/s]
Loading 0:  15%|█▌        | 44/291 [00:00<00:03, 81.62it/s]
Loading 0:  18%|█▊        | 53/291 [00:00<00:02, 81.79it/s]
Loading 0:  21%|██▏       | 62/291 [00:00<00:02, 83.66it/s]
Loading 0:  26%|██▌       | 76/291 [00:00<00:02, 88.08it/s]
Loading 0:  29%|██▉       | 85/291 [00:02<00:09, 21.87it/s]
Loading 0:  33%|███▎      | 95/291 [00:02<00:06, 28.75it/s]
Loading 0:  36%|███▌      | 105/291 [00:02<00:05, 36.67it/s]
Loading 0:  39%|███▉      | 114/291 [00:02<00:04, 43.71it/s]
Loading 0:  42%|████▏     | 123/291 [00:02<00:03, 50.93it/s]
Loading 0:  45%|████▌     | 132/291 [00:02<00:02, 56.77it/s]
Loading 0:  48%|████▊     | 141/291 [00:02<00:02, 61.59it/s]
Loading 0:  52%|█████▏    | 150/291 [00:02<00:02, 63.93it/s]
Loading 0:  54%|█████▍    | 158/291 [00:03<00:01, 67.32it/s]
Loading 0:  57%|█████▋    | 166/291 [00:03<00:01, 70.16it/s]
Loading 0:  61%|██████    | 177/291 [00:03<00:01, 80.29it/s]
Loading 0:  64%|██████▍   | 186/291 [00:03<00:01, 82.04it/s]
Loading 0:  67%|██████▋   | 195/291 [00:04<00:04, 20.89it/s]
Loading 0:  70%|███████   | 204/291 [00:04<00:03, 27.03it/s]
Loading 0:  73%|███████▎  | 212/291 [00:04<00:02, 32.79it/s]
Loading 0:  76%|███████▌  | 220/291 [00:04<00:01, 38.95it/s]
Loading 0:  79%|███████▉  | 231/291 [00:04<00:01, 50.42it/s]
Loading 0:  83%|████████▎ | 241/291 [00:05<00:00, 54.58it/s]
Loading 0:  86%|████████▌ | 250/291 [00:05<00:00, 57.56it/s]
Loading 0:  89%|████████▉ | 259/291 [00:05<00:00, 63.75it/s]
Loading 0:  92%|█████████▏| 269/291 [00:05<00:00, 71.92it/s]
Loading 0:  96%|█████████▌| 278/291 [00:05<00:00, 76.01it/s]
Loading 0:  99%|█████████▊| 287/291 [00:06<00:00, 40.86it/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.
jic062-instruct-fail-v2-mkmlizer:   warnings.warn(
jic062-instruct-fail-v2-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.
jic062-instruct-fail-v2-mkmlizer:   warnings.warn(
jic062-instruct-fail-v2-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.
jic062-instruct-fail-v2-mkmlizer:   warnings.warn(
jic062-instruct-fail-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jic062-instruct-fail-v2-mkmlizer: Saving duration: 1.409s
jic062-instruct-fail-v2-mkmlizer: Processed model Jellywibble/gpt2_xl_pairwise_89m_step_347634 in 11.097s
jic062-instruct-fail-v2-mkmlizer: creating bucket guanaco-reward-models
jic062-instruct-fail-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jic062-instruct-fail-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jic062-instruct-fail-v2_reward
jic062-instruct-fail-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jic062-instruct-fail-v2_reward/config.json
jic062-instruct-fail-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jic062-instruct-fail-v2_reward/special_tokens_map.json
jic062-instruct-fail-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jic062-instruct-fail-v2_reward/tokenizer_config.json
jic062-instruct-fail-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jic062-instruct-fail-v2_reward/merges.txt
jic062-instruct-fail-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jic062-instruct-fail-v2_reward/vocab.json
jic062-instruct-fail-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jic062-instruct-fail-v2_reward/tokenizer.json
jic062-instruct-fail-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/jic062-instruct-fail-v2_reward/reward.tensors
Job jic062-instruct-fail-v2-mkmlizer completed after 95.07s with status: succeeded
Stopping job with name jic062-instruct-fail-v2-mkmlizer
Pipeline stage MKMLizer completed in 96.01s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service jic062-instruct-fail-v2
Waiting for inference service jic062-instruct-fail-v2 to be ready
Inference service jic062-instruct-fail-v2 ready after 191.2196991443634s
Pipeline stage ISVCDeployer completed in 193.21s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1920676231384277s
Received healthy response to inference request in 1.4417104721069336s
Received healthy response to inference request in 1.4428515434265137s
Received healthy response to inference request in 1.4339590072631836s
Received healthy response to inference request in 1.5326578617095947s
5 requests
0 failed requests
5th percentile: 1.4355093002319337
10th percentile: 1.4370595932006835
20th percentile: 1.4401601791381835
30th percentile: 1.4419386863708497
40th percentile: 1.4423951148986816
50th percentile: 1.4428515434265137
60th percentile: 1.478774070739746
70th percentile: 1.5146965980529785
80th percentile: 1.6645398139953613
90th percentile: 1.9283037185668945
95th percentile: 2.060185670852661
99th percentile: 2.1656912326812745
mean time: 1.6086493015289307
Pipeline stage StressChecker completed in 8.72s
jic062-instruct-fail_v2 status is now deployed due to DeploymentManager action
jic062-instruct-fail_v2 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of jic062-instruct-fail_v2
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLDeleter
Checking if service jic062-instruct-fail-v2 is running
admin requested tearing down of jic062-instruct_v16
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLDeleter
Checking if service jic062-instruct-v16 is running
Tearing down inference service jic062-instruct-fail-v2
Connection pool is full, discarding connection: %s. Connection pool size: %s
Service jic062-instruct-fail-v2 has been torndown
Pipeline stage MKMLDeleter completed in 5.08s
run pipeline stage %s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key jic062-instruct-fail-v2/config.json from bucket guanaco-mkml-models
Deleting key jic062-instruct-fail-v2/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Tearing down inference service jic062-instruct-v16
Service jic062-instruct-v16 has been torndown
Deleting key jic062-instruct-fail-v2/special_tokens_map.json from bucket guanaco-mkml-models
Pipeline stage MKMLDeleter completed in 4.01s
Deleting key jic062-instruct-fail-v2/tokenizer.json from bucket guanaco-mkml-models
run pipeline stage %s
Deleting key jic062-instruct-fail-v2/tokenizer_config.json from bucket guanaco-mkml-models
Running pipeline stage MKMLModelDeleter
Cleaning model data from model cache
Cleaning model data from S3
Deleting key jic062-instruct-fail-v2_reward/config.json from bucket guanaco-reward-models
Cleaning model data from model cache
Deleting key jic062-instruct-fail-v2_reward/merges.txt from bucket guanaco-reward-models
Deleting key jic062-instruct-v16/config.json from bucket guanaco-mkml-models
Deleting key jic062-instruct-fail-v2_reward/reward.tensors from bucket guanaco-reward-models
Deleting key jic062-instruct-v16/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key jic062-instruct-fail-v2_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key jic062-instruct-fail-v2_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key jic062-instruct-fail-v2_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key jic062-instruct-v16/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key jic062-instruct-fail-v2_reward/vocab.json from bucket guanaco-reward-models
Deleting key jic062-instruct-v16/tokenizer.json from bucket guanaco-mkml-models
Pipeline stage MKMLModelDeleter completed in 4.56s
Deleting key jic062-instruct-v16/tokenizer_config.json from bucket guanaco-mkml-models
Deleting key jic062-instruct-v16/tokenizer_config.json from bucket guanaco-mkml-models
jic062-instruct-fail_v2 status is now torndown due to DeploymentManager action