submission_id: function_dimum_2024-08-15
developer_uid: chai_backend_admin
alignment_samples: 4116
alignment_score: 2.544263873105642
celo_rating: 1222.92
display_name: gpt4-tl
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': 0.1, 'top_k': 100, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', 'You:'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64, 'reward_max_token_input': 256}
is_internal_developer: True
model_group:
model_name: gpt4-tl
num_battles: 4116
num_wins: 1893
propriety_score: 0.7627118644067796
propriety_total_count: 413.0
ranking_group: single
reward_repo: ChaiML/gpt2_xl_pairwise_89m_step_347634
status: torndown
submission_type: function
timestamp: 2024-08-15T22:50:02+00:00
us_pacific_date: 2024-08-15
win_ratio: 0.45991253644314867
Download Preferencedata
Resubmit model
Running pipeline stage StressChecker
HTTP Request: %s %s "%s %d %s"
Received healthy response to inference request in 1.6399116516113281s
HTTP Request: %s %s "%s %d %s"
Received healthy response to inference request in 1.8589394092559814s
HTTP Request: %s %s "%s %d %s"
Received healthy response to inference request in 1.7790780067443848s
HTTP Request: %s %s "%s %d %s"
Received healthy response to inference request in 1.5559771060943604s
HTTP Request: %s %s "%s %d %s"
Received healthy response to inference request in 1.8820316791534424s
5 requests
0 failed requests
5th percentile: 1.572764015197754
10th percentile: 1.5895509243011474
20th percentile: 1.6231247425079345
30th percentile: 1.6677449226379395
40th percentile: 1.723411464691162
50th percentile: 1.7790780067443848
60th percentile: 1.8110225677490235
70th percentile: 1.8429671287536622
80th percentile: 1.8635578632354737
90th percentile: 1.872794771194458
95th percentile: 1.8774132251739502
99th percentile: 1.881107988357544
mean time: 1.7431875705718993
Pipeline stage StressChecker completed in 11.99s
function_dimum_2024-08-15 status is now deployed due to DeploymentManager action
function_dimum_2024-08-15 status is now inactive due to admin request
function_dimum_2024-08-15 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics