training summary
Browse files- training_log.json +27 -0
training_log.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model": "Qwen/Qwen2.5-7B-Instruct",
|
| 3 |
+
"hub_repo": "pranav-1100/phonepilot-qwen7b",
|
| 4 |
+
"config": {
|
| 5 |
+
"max_seq_len": 2048,
|
| 6 |
+
"lora_r": 16,
|
| 7 |
+
"sft_epochs": 2,
|
| 8 |
+
"max_grpo_steps": 80,
|
| 9 |
+
"grpo_num_generations": 2
|
| 10 |
+
},
|
| 11 |
+
"sft": {
|
| 12 |
+
"phase": "sft",
|
| 13 |
+
"skipped": false,
|
| 14 |
+
"seconds": 1157.6227622032166,
|
| 15 |
+
"episodes": 275,
|
| 16 |
+
"files": 9,
|
| 17 |
+
"sft_sample": "```json\n{\"tool\":\"send_whatsapp\",\"contact\":\"Ria\",\"text\":\"Just letting you know, I'll be 10 mins late for our 4pm meeting.\"}\n```",
|
| 18 |
+
"sft_parses": true
|
| 19 |
+
},
|
| 20 |
+
"grpo": {
|
| 21 |
+
"phase": "grpo",
|
| 22 |
+
"skipped": false,
|
| 23 |
+
"status": "success",
|
| 24 |
+
"seconds": 1959.0953307151794,
|
| 25 |
+
"max_steps": 80
|
| 26 |
+
}
|
| 27 |
+
}
|