developer_uid: chai_backend_admin
submission_id: jellywibble-lora-120k-pr_2801_v4
model_name: jellywibble-lora-120k-pr_2801_v4
model_group: Jellywibble/lora_120k_pr
status: torndown
timestamp: 2024-07-11T20:24:53+00:00
num_battles: 35206
num_wins: 22586
celo_rating: 1292.06
family_friendly_score: 0.0
submission_type: basic
model_repo: Jellywibble/lora_120k_pref_data_ep3_stacked_elo_alignment
model_architecture: LlamaForCausalLM
reward_repo: ChaiML/gpt2_xl_pairwise_89m_step_347634
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 96
display_name: jellywibble-lora-120k-pr_2801_v4
ineligible_reason: max_output_tokens!=64
is_internal_developer: True
language_model: Jellywibble/lora_120k_pref_data_ep3_stacked_elo_alignment
model_size: 8B
ranking_group: single
us_pacific_date: 2024-07-11
win_ratio: 0.6415383741407714
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.08, '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': 96}
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: {message}\n', 'memory_template': '""', 'prompt_template': '""', 'response_template': 'Bot:', 'truncate_by_message': False, 'user_template': 'User: {message}\n'}
Resubmit model
Running pipeline stage MKMLizer
Starting job with name jellywibble-lora-120k-pr-2801-v4-mkmlizer
Waiting for job on jellywibble-lora-120k-pr-2801-v4-mkmlizer to finish
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ _____ __ __ ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ /___/ ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ Version: 0.8.14 ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ https://mk1.ai ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ The license key for the current software has been verified as ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ belonging to: ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ Chai Research Corp. ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ║ ║
jellywibble-lora-120k-pr-2801-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jellywibble-lora-120k-pr-2801-v4-mkmlizer: Downloaded to shared memory in 46.591s
jellywibble-lora-120k-pr-2801-v4-mkmlizer: quantizing model to /dev/shm/model_cache
jellywibble-lora-120k-pr-2801-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jellywibble-lora-120k-pr-2801-v4-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 2%|▏ | 5/291 [00:00<00:06, 43.46it/s] Loading 0: 4%|▍ | 13/291 [00:00<00:04, 62.97it/s] Loading 0: 7%|▋ | 21/291 [00:00<00:03, 69.77it/s] Loading 0: 10%|▉ | 29/291 [00:00<00:03, 72.19it/s] Loading 0: 13%|█▎ | 37/291 [00:00<00:08, 29.56it/s] Loading 0: 15%|█▍ | 43/291 [00:01<00:07, 33.76it/s] Loading 0: 17%|█▋ | 50/291 [00:01<00:06, 38.03it/s] Loading 0: 20%|█▉ | 58/291 [00:01<00:05, 46.23it/s] Loading 0: 23%|██▎ | 66/291 [00:01<00:04, 53.62it/s] Loading 0: 25%|██▌ | 73/291 [00:01<00:03, 57.20it/s] Loading 0: 27%|██▋ | 80/291 [00:02<00:06, 30.59it/s] Loading 0: 30%|██▉ | 86/291 [00:02<00:06, 33.83it/s] Loading 0: 32%|███▏ | 92/291 [00:02<00:05, 36.92it/s] Loading 0: 34%|███▍ | 100/291 [00:02<00:04, 45.15it/s] Loading 0: 37%|███▋ | 108/291 [00:02<00:03, 52.59it/s] Loading 0: 40%|███▉ | 115/291 [00:02<00:03, 56.41it/s] Loading 0: 42%|████▏ | 122/291 [00:02<00:02, 59.49it/s] Loading 0: 44%|████▍ | 129/291 [00:02<00:02, 60.00it/s] Loading 0: 47%|████▋ | 136/291 [00:03<00:05, 30.89it/s] Loading 0: 48%|████▊ | 141/291 [00:03<00:04, 32.65it/s] Loading 0: 51%|█████ | 148/291 [00:03<00:03, 38.38it/s] Loading 0: 54%|█████▎ | 156/291 [00:03<00:03, 44.73it/s] Loading 0: 56%|█████▌ | 162/291 [00:03<00:02, 46.74it/s] Loading 0: 58%|█████▊ | 168/291 [00:03<00:02, 46.93it/s] Loading 0: 60%|██████ | 175/291 [00:03<00:02, 49.66it/s] Loading 0: 63%|██████▎ | 183/291 [00:04<00:01, 57.01it/s] Loading 0: 65%|██████▌ | 190/291 [00:04<00:03, 29.05it/s] Loading 0: 67%|██████▋ | 195/291 [00:04<00:03, 31.43it/s] Loading 0: 69%|██████▉ | 202/291 [00:04<00:02, 38.07it/s] Loading 0: 72%|███████▏ | 210/291 [00:04<00:01, 45.33it/s] Loading 0: 75%|███████▍ | 217/291 [00:05<00:01, 50.50it/s] Loading 0: 77%|███████▋ | 224/291 [00:05<00:01, 51.71it/s] Loading 0: 79%|███████▉ | 230/291 [00:05<00:01, 53.45it/s] Loading 0: 81%|████████ | 236/291 [00:05<00:01, 30.84it/s] Loading 0: 83%|████████▎ | 241/291 [00:05<00:01, 33.36it/s] Loading 0: 85%|████████▌ | 248/291 [00:05<00:01, 38.04it/s] Loading 0: 88%|████████▊ | 256/291 [00:06<00:00, 46.34it/s] Loading 0: 91%|█████████ | 264/291 [00:06<00:00, 53.64it/s] Loading 0: 94%|█████████▍| 273/291 [00:06<00:00, 57.53it/s] Loading 0: 97%|█████████▋| 281/291 [00:06<00:00, 58.03it/s] Loading 0: 99%|█████████▉| 288/291 [00:13<00:00, 3.43it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
jellywibble-lora-120k-pr-2801-v4-mkmlizer: quantized model in 32.647s
jellywibble-lora-120k-pr-2801-v4-mkmlizer: Processed model Jellywibble/lora_120k_pref_data_ep3_stacked_elo_alignment in 79.238s
jellywibble-lora-120k-pr-2801-v4-mkmlizer: creating bucket guanaco-mkml-models
jellywibble-lora-120k-pr-2801-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jellywibble-lora-120k-pr-2801-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2801-v4
jellywibble-lora-120k-pr-2801-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2801-v4/special_tokens_map.json
jellywibble-lora-120k-pr-2801-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2801-v4/tokenizer_config.json
jellywibble-lora-120k-pr-2801-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2801-v4/tokenizer.json
jellywibble-lora-120k-pr-2801-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2801-v4/config.json
jellywibble-lora-120k-pr-2801-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2801-v4/flywheel_model.0.safetensors
jellywibble-lora-120k-pr-2801-v4-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
jellywibble-lora-120k-pr-2801-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.
jellywibble-lora-120k-pr-2801-v4-mkmlizer: warnings.warn(
jellywibble-lora-120k-pr-2801-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`.
jellywibble-lora-120k-pr-2801-v4-mkmlizer: warnings.warn(
jellywibble-lora-120k-pr-2801-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.
jellywibble-lora-120k-pr-2801-v4-mkmlizer: warnings.warn(
jellywibble-lora-120k-pr-2801-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.
jellywibble-lora-120k-pr-2801-v4-mkmlizer: warnings.warn(
jellywibble-lora-120k-pr-2801-v4-mkmlizer: Saving duration: 2.185s
jellywibble-lora-120k-pr-2801-v4-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 14.122s
jellywibble-lora-120k-pr-2801-v4-mkmlizer: creating bucket guanaco-reward-models
jellywibble-lora-120k-pr-2801-v4-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jellywibble-lora-120k-pr-2801-v4-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jellywibble-lora-120k-pr-2801-v4_reward
jellywibble-lora-120k-pr-2801-v4-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2801-v4_reward/special_tokens_map.json
jellywibble-lora-120k-pr-2801-v4-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2801-v4_reward/tokenizer_config.json
jellywibble-lora-120k-pr-2801-v4-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2801-v4_reward/vocab.json
jellywibble-lora-120k-pr-2801-v4-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2801-v4_reward/config.json
jellywibble-lora-120k-pr-2801-v4-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jellywibble-lora-120k-pr-2801-v4_reward/merges.txt
jellywibble-lora-120k-pr-2801-v4-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2801-v4_reward/tokenizer.json
jellywibble-lora-120k-pr-2801-v4-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/jellywibble-lora-120k-pr-2801-v4_reward/reward.tensors
Job jellywibble-lora-120k-pr-2801-v4-mkmlizer completed after 128.64s with status: succeeded
Stopping job with name jellywibble-lora-120k-pr-2801-v4-mkmlizer
Pipeline stage MKMLizer completed in 129.64s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service jellywibble-lora-120k-pr-2801-v4
Connection pool is full, discarding connection: %s
Waiting for inference service jellywibble-lora-120k-pr-2801-v4 to be ready
Inference service jellywibble-lora-120k-pr-2801-v4 ready after 40.198503255844116s
Pipeline stage ISVCDeployer completed in 47.57s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.5479564666748047s
Received healthy response to inference request in 1.8835265636444092s
Received healthy response to inference request in 1.8551721572875977s
Received healthy response to inference request in 1.8299386501312256s
Received healthy response to inference request in 1.9290387630462646s
5 requests
0 failed requests
5th percentile: 1.8349853515625
10th percentile: 1.8400320529937744
20th percentile: 1.8501254558563232
30th percentile: 1.86084303855896
40th percentile: 1.8721848011016846
50th percentile: 1.8835265636444092
60th percentile: 1.9017314434051513
70th percentile: 1.9199363231658935
80th percentile: 2.052822303771973
90th percentile: 2.3003893852233888
95th percentile: 2.4241729259490965
99th percentile: 2.523199758529663
mean time: 2.0091265201568604
Pipeline stage StressChecker completed in 10.83s
jellywibble-lora-120k-pr_2801_v4 status is now deployed due to DeploymentManager action
jellywibble-lora-120k-pr_2801_v4 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of jellywibble-lora-120k-pr_2801_v4
Running pipeline stage ISVCDeleter
Checking if service jellywibble-lora-120k-pr-2801-v4 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 4.16s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key jellywibble-lora-120k-pr-2801-v4/config.json from bucket guanaco-mkml-models
Deleting key jellywibble-lora-120k-pr-2801-v4/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key jellywibble-lora-120k-pr-2801-v4/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key jellywibble-lora-120k-pr-2801-v4/tokenizer.json from bucket guanaco-mkml-models
Deleting key jellywibble-lora-120k-pr-2801-v4/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key jellywibble-lora-120k-pr-2801-v4_reward/config.json from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2801-v4_reward/merges.txt from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2801-v4_reward/reward.tensors from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2801-v4_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2801-v4_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2801-v4_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2801-v4_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.45s
jellywibble-lora-120k-pr_2801_v4 status is now torndown due to DeploymentManager action