submission_id: function_negir_2024-08-09
developer_uid: chai_backend_admin
alignment_samples: 0
display_name: gpt-3-5
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}
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'], 'max_input_tokens': 512, 'best_of': 1, 'max_output_tokens': 64, 'reward_max_token_input': 256}
is_internal_developer: True
model_group:
model_name: gpt-3-5
num_battles: 228
num_wins: 67
propriety_score: 0.8095238095238095
propriety_total_count: 21.0
ranking_group: single
reward_repo: ChaiML/gpt2_xl_pairwise_89m_step_347634
status: torndown
submission_type: function
timestamp: 2024-08-09T18:24:39+00:00
us_pacific_date: 2024-08-09
win_ratio: 0.29385964912280704
Download Preference Data
Resubmit model
Running pipeline stage StressChecker
Received healthy response to inference request in 0.5771605968475342s
Received healthy response to inference request in 0.40593934059143066s
Received healthy response to inference request in 0.4492063522338867s
Received healthy response to inference request in 0.39270782470703125s
Received healthy response to inference request in 0.46718931198120117s
5 requests
0 failed requests
5th percentile: 0.3953541278839111
10th percentile: 0.398000431060791
20th percentile: 0.4032930374145508
30th percentile: 0.4145927429199219
40th percentile: 0.4318995475769043
50th percentile: 0.4492063522338867
60th percentile: 0.4563995361328125
70th percentile: 0.4635927200317383
80th percentile: 0.4891835689544678
90th percentile: 0.5331720829010009
95th percentile: 0.5551663398742676
99th percentile: 0.5727617454528808
mean time: 0.4584406852722168
Pipeline stage StressChecker completed in 2.89s
function_negir_2024-08-09 status is now deployed due to DeploymentManager action
function_negir_2024-08-09 status is now torndown due to DeploymentManager action
function_negir_2024-08-09 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics