chore(receipt): chaski-5050 unique Unsloth MEASURED loss
Browse files- training_receipt.unsloth.json +15 -12
training_receipt.unsloth.json
CHANGED
|
@@ -1,19 +1,22 @@
|
|
| 1 |
{
|
| 2 |
"schema": "szl.training_receipt.v2",
|
| 3 |
-
"profile": "chaski",
|
| 4 |
-
"artifact": "SZLHOLDINGS/chaski",
|
| 5 |
"base_model": "Qwen/Qwen3.5-0.8B",
|
| 6 |
"base_model_runtime": "unsloth/Qwen3.5-0.8B",
|
| 7 |
-
"cut": "
|
| 8 |
"lora": {
|
| 9 |
-
"r":
|
| 10 |
"alpha": 16,
|
| 11 |
"rslora": true,
|
| 12 |
"targets": [
|
| 13 |
"q_proj",
|
| 14 |
"k_proj",
|
| 15 |
"v_proj",
|
| 16 |
-
"o_proj"
|
|
|
|
|
|
|
|
|
|
| 17 |
],
|
| 18 |
"dropout": 0,
|
| 19 |
"bias": "none",
|
|
@@ -34,13 +37,13 @@
|
|
| 34 |
"dataset_sha256": "ddc5594bfb1c78449ba40a263f5ac41d21c896c3c7ed7346341c7c080611a243",
|
| 35 |
"training_rows": 45,
|
| 36 |
"seed": 20260721,
|
| 37 |
-
"training_loss":
|
| 38 |
"metrics": {
|
| 39 |
-
"train_runtime":
|
| 40 |
-
"train_samples_per_second": 3.
|
| 41 |
-
"train_steps_per_second": 0.
|
| 42 |
-
"total_flos":
|
| 43 |
-
"train_loss":
|
| 44 |
"epoch": 20.0
|
| 45 |
},
|
| 46 |
"honesty": "MEASURED",
|
|
@@ -55,5 +58,5 @@
|
|
| 55 |
"path_in_repo": "adapter-unsloth",
|
| 56 |
"lambda": "Conjecture 1",
|
| 57 |
"doctrine": "v11 LOCKED 749/14/163",
|
| 58 |
-
"computed_at": "2026-08-29T18:
|
| 59 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"schema": "szl.training_receipt.v2",
|
| 3 |
+
"profile": "chaski-5050",
|
| 4 |
+
"artifact": "SZLHOLDINGS/chaski-5050",
|
| 5 |
"base_model": "Qwen/Qwen3.5-0.8B",
|
| 6 |
"base_model_runtime": "unsloth/Qwen3.5-0.8B",
|
| 7 |
+
"cut": "50/50 cutting mix. Extra MLP rank so the courier is allowed to STOP.",
|
| 8 |
"lora": {
|
| 9 |
+
"r": 16,
|
| 10 |
"alpha": 16,
|
| 11 |
"rslora": true,
|
| 12 |
"targets": [
|
| 13 |
"q_proj",
|
| 14 |
"k_proj",
|
| 15 |
"v_proj",
|
| 16 |
+
"o_proj",
|
| 17 |
+
"gate_proj",
|
| 18 |
+
"up_proj",
|
| 19 |
+
"down_proj"
|
| 20 |
],
|
| 21 |
"dropout": 0,
|
| 22 |
"bias": "none",
|
|
|
|
| 37 |
"dataset_sha256": "ddc5594bfb1c78449ba40a263f5ac41d21c896c3c7ed7346341c7c080611a243",
|
| 38 |
"training_rows": 45,
|
| 39 |
"seed": 20260721,
|
| 40 |
+
"training_loss": 0.5588374246212576,
|
| 41 |
"metrics": {
|
| 42 |
+
"train_runtime": 302.416,
|
| 43 |
+
"train_samples_per_second": 3.174,
|
| 44 |
+
"train_steps_per_second": 0.397,
|
| 45 |
+
"total_flos": 327704043601536.0,
|
| 46 |
+
"train_loss": 0.5588374246212576,
|
| 47 |
"epoch": 20.0
|
| 48 |
},
|
| 49 |
"honesty": "MEASURED",
|
|
|
|
| 58 |
"path_in_repo": "adapter-unsloth",
|
| 59 |
"lambda": "Conjecture 1",
|
| 60 |
"doctrine": "v11 LOCKED 749/14/163",
|
| 61 |
+
"computed_at": "2026-08-29T18:43:09.568494+00:00"
|
| 62 |
}
|