Commit
·
0982c6d
1
Parent(s):
d787fb4
initial release v3
Browse files- adapter_config.json +27 -0
- adapter_model.bin +3 -0
- training_graph.json +956 -0
- training_graph.png +0 -0
- training_log.json +19 -0
- training_parameters.json +37 -0
- training_prompt.json +3 -0
adapter_config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "models/meta-llama_Meta-Llama-3-8B",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": {},
|
| 12 |
+
"lora_alpha": 256,
|
| 13 |
+
"lora_dropout": 0.05,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 128,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": null,
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"v_proj",
|
| 23 |
+
"q_proj"
|
| 24 |
+
],
|
| 25 |
+
"task_type": "CAUSAL_LM",
|
| 26 |
+
"use_rslora": false
|
| 27 |
+
}
|
adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5055b89b8fe192656f23926b31901bbb81d0bb979f6566f1d692fffc5998b01f
|
| 3 |
+
size 218150282
|
training_graph.json
ADDED
|
@@ -0,0 +1,956 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"current_steps": 63,
|
| 4 |
+
"loss": 2.8305,
|
| 5 |
+
"learning_rate": 0.0005,
|
| 6 |
+
"epoch": 0.02549800796812749
|
| 7 |
+
},
|
| 8 |
+
{
|
| 9 |
+
"current_steps": 127,
|
| 10 |
+
"loss": 2.7784,
|
| 11 |
+
"learning_rate": 0.001,
|
| 12 |
+
"epoch": 0.05099601593625498
|
| 13 |
+
},
|
| 14 |
+
{
|
| 15 |
+
"current_steps": 191,
|
| 16 |
+
"loss": 2.6623,
|
| 17 |
+
"learning_rate": 0.0009974226804123712,
|
| 18 |
+
"epoch": 0.07649402390438247
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"current_steps": 255,
|
| 22 |
+
"loss": 2.6442,
|
| 23 |
+
"learning_rate": 0.0009948453608247423,
|
| 24 |
+
"epoch": 0.10199203187250996
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"current_steps": 319,
|
| 28 |
+
"loss": 2.822,
|
| 29 |
+
"learning_rate": 0.0009922680412371135,
|
| 30 |
+
"epoch": 0.12749003984063745
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"current_steps": 383,
|
| 34 |
+
"loss": 2.6869,
|
| 35 |
+
"learning_rate": 0.0009896907216494846,
|
| 36 |
+
"epoch": 0.15298804780876493
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"current_steps": 447,
|
| 40 |
+
"loss": 2.5865,
|
| 41 |
+
"learning_rate": 0.0009871134020618558,
|
| 42 |
+
"epoch": 0.17848605577689244
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"current_steps": 511,
|
| 46 |
+
"loss": 2.6222,
|
| 47 |
+
"learning_rate": 0.000984536082474227,
|
| 48 |
+
"epoch": 0.20398406374501993
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"current_steps": 575,
|
| 52 |
+
"loss": 2.5421,
|
| 53 |
+
"learning_rate": 0.0009819587628865979,
|
| 54 |
+
"epoch": 0.2294820717131474
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"current_steps": 639,
|
| 58 |
+
"loss": 2.5766,
|
| 59 |
+
"learning_rate": 0.000979381443298969,
|
| 60 |
+
"epoch": 0.2549800796812749
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"current_steps": 703,
|
| 64 |
+
"loss": 2.5266,
|
| 65 |
+
"learning_rate": 0.0009768041237113402,
|
| 66 |
+
"epoch": 0.2804780876494024
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"current_steps": 767,
|
| 70 |
+
"loss": 2.5239,
|
| 71 |
+
"learning_rate": 0.0009742268041237113,
|
| 72 |
+
"epoch": 0.30597609561752986
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"current_steps": 831,
|
| 76 |
+
"loss": 2.4799,
|
| 77 |
+
"learning_rate": 0.0009716494845360825,
|
| 78 |
+
"epoch": 0.3314741035856574
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"current_steps": 895,
|
| 82 |
+
"loss": 2.5332,
|
| 83 |
+
"learning_rate": 0.0009690721649484536,
|
| 84 |
+
"epoch": 0.3569721115537849
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"current_steps": 959,
|
| 88 |
+
"loss": 2.4283,
|
| 89 |
+
"learning_rate": 0.0009664948453608248,
|
| 90 |
+
"epoch": 0.38247011952191234
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"current_steps": 1023,
|
| 94 |
+
"loss": 2.4595,
|
| 95 |
+
"learning_rate": 0.0009639175257731959,
|
| 96 |
+
"epoch": 0.40796812749003986
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"current_steps": 1087,
|
| 100 |
+
"loss": 2.4821,
|
| 101 |
+
"learning_rate": 0.0009613402061855671,
|
| 102 |
+
"epoch": 0.4334661354581673
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"current_steps": 1151,
|
| 106 |
+
"loss": 2.4473,
|
| 107 |
+
"learning_rate": 0.0009587628865979382,
|
| 108 |
+
"epoch": 0.4589641434262948
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"current_steps": 1215,
|
| 112 |
+
"loss": 2.4614,
|
| 113 |
+
"learning_rate": 0.0009561855670103094,
|
| 114 |
+
"epoch": 0.48446215139442234
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"current_steps": 1279,
|
| 118 |
+
"loss": 2.4037,
|
| 119 |
+
"learning_rate": 0.0009536082474226805,
|
| 120 |
+
"epoch": 0.5099601593625498
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"current_steps": 1343,
|
| 124 |
+
"loss": 2.4243,
|
| 125 |
+
"learning_rate": 0.0009510309278350515,
|
| 126 |
+
"epoch": 0.5354581673306773
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"current_steps": 1407,
|
| 130 |
+
"loss": 2.3534,
|
| 131 |
+
"learning_rate": 0.0009484536082474226,
|
| 132 |
+
"epoch": 0.5609561752988048
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"current_steps": 1471,
|
| 136 |
+
"loss": 2.406,
|
| 137 |
+
"learning_rate": 0.0009458762886597938,
|
| 138 |
+
"epoch": 0.5864541832669322
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"current_steps": 1535,
|
| 142 |
+
"loss": 2.483,
|
| 143 |
+
"learning_rate": 0.0009432989690721649,
|
| 144 |
+
"epoch": 0.6119521912350597
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"current_steps": 1599,
|
| 148 |
+
"loss": 2.4313,
|
| 149 |
+
"learning_rate": 0.0009407216494845361,
|
| 150 |
+
"epoch": 0.6374501992031872
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"current_steps": 1663,
|
| 154 |
+
"loss": 2.4101,
|
| 155 |
+
"learning_rate": 0.0009381443298969072,
|
| 156 |
+
"epoch": 0.6629482071713148
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"current_steps": 1727,
|
| 160 |
+
"loss": 2.3847,
|
| 161 |
+
"learning_rate": 0.0009355670103092784,
|
| 162 |
+
"epoch": 0.6884462151394423
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"current_steps": 1791,
|
| 166 |
+
"loss": 2.3194,
|
| 167 |
+
"learning_rate": 0.0009329896907216495,
|
| 168 |
+
"epoch": 0.7139442231075698
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"current_steps": 1855,
|
| 172 |
+
"loss": 2.3921,
|
| 173 |
+
"learning_rate": 0.0009304123711340207,
|
| 174 |
+
"epoch": 0.7394422310756972
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"current_steps": 1919,
|
| 178 |
+
"loss": 2.4158,
|
| 179 |
+
"learning_rate": 0.0009278350515463918,
|
| 180 |
+
"epoch": 0.7649402390438247
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"current_steps": 1983,
|
| 184 |
+
"loss": 2.3134,
|
| 185 |
+
"learning_rate": 0.000925257731958763,
|
| 186 |
+
"epoch": 0.7904382470119522
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"current_steps": 2047,
|
| 190 |
+
"loss": 2.3514,
|
| 191 |
+
"learning_rate": 0.0009226804123711341,
|
| 192 |
+
"epoch": 0.8159362549800797
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"current_steps": 2111,
|
| 196 |
+
"loss": 2.3011,
|
| 197 |
+
"learning_rate": 0.0009201030927835051,
|
| 198 |
+
"epoch": 0.8414342629482072
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"current_steps": 2175,
|
| 202 |
+
"loss": 2.4016,
|
| 203 |
+
"learning_rate": 0.0009175257731958762,
|
| 204 |
+
"epoch": 0.8669322709163346
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"current_steps": 2239,
|
| 208 |
+
"loss": 2.3209,
|
| 209 |
+
"learning_rate": 0.0009149484536082474,
|
| 210 |
+
"epoch": 0.8924302788844621
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"current_steps": 2303,
|
| 214 |
+
"loss": 2.3672,
|
| 215 |
+
"learning_rate": 0.0009123711340206185,
|
| 216 |
+
"epoch": 0.9179282868525896
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"current_steps": 2367,
|
| 220 |
+
"loss": 2.3597,
|
| 221 |
+
"learning_rate": 0.0009097938144329897,
|
| 222 |
+
"epoch": 0.9434262948207172
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"current_steps": 2431,
|
| 226 |
+
"loss": 2.3619,
|
| 227 |
+
"learning_rate": 0.0009072164948453608,
|
| 228 |
+
"epoch": 0.9689243027888447
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"current_steps": 2495,
|
| 232 |
+
"loss": 2.3521,
|
| 233 |
+
"learning_rate": 0.000904639175257732,
|
| 234 |
+
"epoch": 0.9944223107569721
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"current_steps": 2545,
|
| 238 |
+
"loss": 2.2467,
|
| 239 |
+
"learning_rate": 0.0009020618556701031,
|
| 240 |
+
"epoch": 1.0199203187250996
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"current_steps": 2609,
|
| 244 |
+
"loss": 2.2082,
|
| 245 |
+
"learning_rate": 0.0008994845360824743,
|
| 246 |
+
"epoch": 1.045418326693227
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"current_steps": 2673,
|
| 250 |
+
"loss": 2.2161,
|
| 251 |
+
"learning_rate": 0.0008969072164948454,
|
| 252 |
+
"epoch": 1.0709163346613546
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"current_steps": 2737,
|
| 256 |
+
"loss": 2.1961,
|
| 257 |
+
"learning_rate": 0.0008943298969072166,
|
| 258 |
+
"epoch": 1.0964143426294821
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"current_steps": 2801,
|
| 262 |
+
"loss": 2.215,
|
| 263 |
+
"learning_rate": 0.0008917525773195877,
|
| 264 |
+
"epoch": 1.1219123505976096
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"current_steps": 2865,
|
| 268 |
+
"loss": 2.1951,
|
| 269 |
+
"learning_rate": 0.0008891752577319587,
|
| 270 |
+
"epoch": 1.1474103585657371
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"current_steps": 2929,
|
| 274 |
+
"loss": 2.1599,
|
| 275 |
+
"learning_rate": 0.0008865979381443298,
|
| 276 |
+
"epoch": 1.1729083665338647
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"current_steps": 2993,
|
| 280 |
+
"loss": 2.1511,
|
| 281 |
+
"learning_rate": 0.000884020618556701,
|
| 282 |
+
"epoch": 1.198406374501992
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"current_steps": 3057,
|
| 286 |
+
"loss": 2.0713,
|
| 287 |
+
"learning_rate": 0.0008814432989690721,
|
| 288 |
+
"epoch": 1.2239043824701195
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"current_steps": 3121,
|
| 292 |
+
"loss": 2.1312,
|
| 293 |
+
"learning_rate": 0.0008788659793814433,
|
| 294 |
+
"epoch": 1.249402390438247
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"current_steps": 3185,
|
| 298 |
+
"loss": 2.1442,
|
| 299 |
+
"learning_rate": 0.0008762886597938144,
|
| 300 |
+
"epoch": 1.2749003984063745
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"current_steps": 3249,
|
| 304 |
+
"loss": 2.141,
|
| 305 |
+
"learning_rate": 0.0008737113402061856,
|
| 306 |
+
"epoch": 1.300398406374502
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"current_steps": 3313,
|
| 310 |
+
"loss": 2.1281,
|
| 311 |
+
"learning_rate": 0.0008711340206185567,
|
| 312 |
+
"epoch": 1.3258964143426295
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"current_steps": 3377,
|
| 316 |
+
"loss": 2.1373,
|
| 317 |
+
"learning_rate": 0.0008685567010309279,
|
| 318 |
+
"epoch": 1.351394422310757
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"current_steps": 3441,
|
| 322 |
+
"loss": 2.1111,
|
| 323 |
+
"learning_rate": 0.000865979381443299,
|
| 324 |
+
"epoch": 1.3768924302788845
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"current_steps": 3505,
|
| 328 |
+
"loss": 2.0973,
|
| 329 |
+
"learning_rate": 0.0008634020618556702,
|
| 330 |
+
"epoch": 1.402390438247012
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"current_steps": 3569,
|
| 334 |
+
"loss": 2.1299,
|
| 335 |
+
"learning_rate": 0.0008608247422680414,
|
| 336 |
+
"epoch": 1.4278884462151393
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"current_steps": 3633,
|
| 340 |
+
"loss": 2.0971,
|
| 341 |
+
"learning_rate": 0.0008582474226804123,
|
| 342 |
+
"epoch": 1.453386454183267
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"current_steps": 3697,
|
| 346 |
+
"loss": 2.0334,
|
| 347 |
+
"learning_rate": 0.0008556701030927834,
|
| 348 |
+
"epoch": 1.4788844621513944
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"current_steps": 3761,
|
| 352 |
+
"loss": 2.0659,
|
| 353 |
+
"learning_rate": 0.0008530927835051546,
|
| 354 |
+
"epoch": 1.5043824701195219
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"current_steps": 3825,
|
| 358 |
+
"loss": 2.0648,
|
| 359 |
+
"learning_rate": 0.0008505154639175257,
|
| 360 |
+
"epoch": 1.5298804780876494
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"current_steps": 3889,
|
| 364 |
+
"loss": 2.0629,
|
| 365 |
+
"learning_rate": 0.0008479381443298969,
|
| 366 |
+
"epoch": 1.5553784860557769
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"current_steps": 3953,
|
| 370 |
+
"loss": 2.0136,
|
| 371 |
+
"learning_rate": 0.000845360824742268,
|
| 372 |
+
"epoch": 1.5808764940239044
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"current_steps": 4017,
|
| 376 |
+
"loss": 2.0692,
|
| 377 |
+
"learning_rate": 0.0008427835051546392,
|
| 378 |
+
"epoch": 1.606374501992032
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"current_steps": 4081,
|
| 382 |
+
"loss": 2.1114,
|
| 383 |
+
"learning_rate": 0.0008402061855670104,
|
| 384 |
+
"epoch": 1.6318725099601594
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"current_steps": 4145,
|
| 388 |
+
"loss": 2.048,
|
| 389 |
+
"learning_rate": 0.0008376288659793815,
|
| 390 |
+
"epoch": 1.6573705179282867
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"current_steps": 4209,
|
| 394 |
+
"loss": 2.1087,
|
| 395 |
+
"learning_rate": 0.0008350515463917527,
|
| 396 |
+
"epoch": 1.6828685258964144
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"current_steps": 4273,
|
| 400 |
+
"loss": 2.0474,
|
| 401 |
+
"learning_rate": 0.0008324742268041238,
|
| 402 |
+
"epoch": 1.7083665338645417
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"current_steps": 4337,
|
| 406 |
+
"loss": 2.0699,
|
| 407 |
+
"learning_rate": 0.000829896907216495,
|
| 408 |
+
"epoch": 1.7338645418326695
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"current_steps": 4401,
|
| 412 |
+
"loss": 2.0901,
|
| 413 |
+
"learning_rate": 0.0008273195876288659,
|
| 414 |
+
"epoch": 1.7593625498007968
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"current_steps": 4465,
|
| 418 |
+
"loss": 2.018,
|
| 419 |
+
"learning_rate": 0.000824742268041237,
|
| 420 |
+
"epoch": 1.7848605577689243
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"current_steps": 4529,
|
| 424 |
+
"loss": 2.039,
|
| 425 |
+
"learning_rate": 0.0008221649484536082,
|
| 426 |
+
"epoch": 1.8103585657370518
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"current_steps": 4593,
|
| 430 |
+
"loss": 2.0168,
|
| 431 |
+
"learning_rate": 0.0008195876288659793,
|
| 432 |
+
"epoch": 1.8358565737051793
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"current_steps": 4657,
|
| 436 |
+
"loss": 1.9646,
|
| 437 |
+
"learning_rate": 0.0008170103092783505,
|
| 438 |
+
"epoch": 1.8613545816733068
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"current_steps": 4721,
|
| 442 |
+
"loss": 1.9875,
|
| 443 |
+
"learning_rate": 0.0008144329896907217,
|
| 444 |
+
"epoch": 1.886852589641434
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"current_steps": 4785,
|
| 448 |
+
"loss": 2.0373,
|
| 449 |
+
"learning_rate": 0.0008118556701030928,
|
| 450 |
+
"epoch": 1.9123505976095618
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"current_steps": 4849,
|
| 454 |
+
"loss": 1.9158,
|
| 455 |
+
"learning_rate": 0.000809278350515464,
|
| 456 |
+
"epoch": 1.9378486055776891
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"current_steps": 4913,
|
| 460 |
+
"loss": 1.9173,
|
| 461 |
+
"learning_rate": 0.0008067010309278351,
|
| 462 |
+
"epoch": 1.9633466135458169
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"current_steps": 4977,
|
| 466 |
+
"loss": 1.9941,
|
| 467 |
+
"learning_rate": 0.0008041237113402063,
|
| 468 |
+
"epoch": 1.9888446215139441
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"current_steps": 5027,
|
| 472 |
+
"loss": 1.8231,
|
| 473 |
+
"learning_rate": 0.0008015463917525774,
|
| 474 |
+
"epoch": 2.014342629482072
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"current_steps": 5091,
|
| 478 |
+
"loss": 1.8018,
|
| 479 |
+
"learning_rate": 0.0007989690721649486,
|
| 480 |
+
"epoch": 2.039840637450199
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"current_steps": 5155,
|
| 484 |
+
"loss": 1.7098,
|
| 485 |
+
"learning_rate": 0.0007963917525773195,
|
| 486 |
+
"epoch": 2.065338645418327
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"current_steps": 5219,
|
| 490 |
+
"loss": 1.7214,
|
| 491 |
+
"learning_rate": 0.0007938144329896907,
|
| 492 |
+
"epoch": 2.090836653386454
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"current_steps": 5283,
|
| 496 |
+
"loss": 1.8029,
|
| 497 |
+
"learning_rate": 0.0007912371134020618,
|
| 498 |
+
"epoch": 2.1163346613545815
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"current_steps": 5347,
|
| 502 |
+
"loss": 1.71,
|
| 503 |
+
"learning_rate": 0.000788659793814433,
|
| 504 |
+
"epoch": 2.141832669322709
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"current_steps": 5411,
|
| 508 |
+
"loss": 1.725,
|
| 509 |
+
"learning_rate": 0.0007860824742268041,
|
| 510 |
+
"epoch": 2.1673306772908365
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"current_steps": 5475,
|
| 514 |
+
"loss": 1.729,
|
| 515 |
+
"learning_rate": 0.0007835051546391753,
|
| 516 |
+
"epoch": 2.1928286852589642
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"current_steps": 5539,
|
| 520 |
+
"loss": 1.7139,
|
| 521 |
+
"learning_rate": 0.0007809278350515464,
|
| 522 |
+
"epoch": 2.2183266932270915
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"current_steps": 5603,
|
| 526 |
+
"loss": 1.6588,
|
| 527 |
+
"learning_rate": 0.0007783505154639176,
|
| 528 |
+
"epoch": 2.2438247011952193
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"current_steps": 5667,
|
| 532 |
+
"loss": 1.7179,
|
| 533 |
+
"learning_rate": 0.0007757731958762887,
|
| 534 |
+
"epoch": 2.2693227091633466
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"current_steps": 5731,
|
| 538 |
+
"loss": 1.7024,
|
| 539 |
+
"learning_rate": 0.0007731958762886599,
|
| 540 |
+
"epoch": 2.2948207171314743
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"current_steps": 5795,
|
| 544 |
+
"loss": 1.7205,
|
| 545 |
+
"learning_rate": 0.000770618556701031,
|
| 546 |
+
"epoch": 2.3203187250996016
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"current_steps": 5859,
|
| 550 |
+
"loss": 1.7094,
|
| 551 |
+
"learning_rate": 0.0007680412371134022,
|
| 552 |
+
"epoch": 2.3458167330677293
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"current_steps": 5923,
|
| 556 |
+
"loss": 1.7152,
|
| 557 |
+
"learning_rate": 0.0007654639175257731,
|
| 558 |
+
"epoch": 2.3713147410358566
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"current_steps": 5987,
|
| 562 |
+
"loss": 1.7285,
|
| 563 |
+
"learning_rate": 0.0007628865979381443,
|
| 564 |
+
"epoch": 2.396812749003984
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"current_steps": 6051,
|
| 568 |
+
"loss": 1.6915,
|
| 569 |
+
"learning_rate": 0.0007603092783505154,
|
| 570 |
+
"epoch": 2.4223107569721116
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"current_steps": 6115,
|
| 574 |
+
"loss": 1.6536,
|
| 575 |
+
"learning_rate": 0.0007577319587628866,
|
| 576 |
+
"epoch": 2.447808764940239
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"current_steps": 6179,
|
| 580 |
+
"loss": 1.6924,
|
| 581 |
+
"learning_rate": 0.0007551546391752577,
|
| 582 |
+
"epoch": 2.4733067729083666
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"current_steps": 6243,
|
| 586 |
+
"loss": 1.6518,
|
| 587 |
+
"learning_rate": 0.0007525773195876289,
|
| 588 |
+
"epoch": 2.498804780876494
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"current_steps": 6307,
|
| 592 |
+
"loss": 1.6765,
|
| 593 |
+
"learning_rate": 0.00075,
|
| 594 |
+
"epoch": 2.5243027888446217
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"current_steps": 6371,
|
| 598 |
+
"loss": 1.6662,
|
| 599 |
+
"learning_rate": 0.0007474226804123712,
|
| 600 |
+
"epoch": 2.549800796812749
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"current_steps": 6435,
|
| 604 |
+
"loss": 1.6179,
|
| 605 |
+
"learning_rate": 0.0007448453608247423,
|
| 606 |
+
"epoch": 2.5752988047808767
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"current_steps": 6499,
|
| 610 |
+
"loss": 1.6485,
|
| 611 |
+
"learning_rate": 0.0007422680412371135,
|
| 612 |
+
"epoch": 2.600796812749004
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"current_steps": 6563,
|
| 616 |
+
"loss": 1.7025,
|
| 617 |
+
"learning_rate": 0.0007396907216494846,
|
| 618 |
+
"epoch": 2.6262948207171313
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"current_steps": 6627,
|
| 622 |
+
"loss": 1.6723,
|
| 623 |
+
"learning_rate": 0.0007371134020618558,
|
| 624 |
+
"epoch": 2.651792828685259
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"current_steps": 6691,
|
| 628 |
+
"loss": 1.7608,
|
| 629 |
+
"learning_rate": 0.0007345360824742269,
|
| 630 |
+
"epoch": 2.6772908366533863
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"current_steps": 6755,
|
| 634 |
+
"loss": 1.6388,
|
| 635 |
+
"learning_rate": 0.0007319587628865979,
|
| 636 |
+
"epoch": 2.702788844621514
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"current_steps": 6819,
|
| 640 |
+
"loss": 1.6381,
|
| 641 |
+
"learning_rate": 0.000729381443298969,
|
| 642 |
+
"epoch": 2.7282868525896413
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"current_steps": 6883,
|
| 646 |
+
"loss": 1.6857,
|
| 647 |
+
"learning_rate": 0.0007268041237113402,
|
| 648 |
+
"epoch": 2.753784860557769
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"current_steps": 6947,
|
| 652 |
+
"loss": 1.6556,
|
| 653 |
+
"learning_rate": 0.0007242268041237113,
|
| 654 |
+
"epoch": 2.7792828685258963
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"current_steps": 7011,
|
| 658 |
+
"loss": 1.6446,
|
| 659 |
+
"learning_rate": 0.0007216494845360825,
|
| 660 |
+
"epoch": 2.804780876494024
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"current_steps": 7075,
|
| 664 |
+
"loss": 1.593,
|
| 665 |
+
"learning_rate": 0.0007190721649484536,
|
| 666 |
+
"epoch": 2.8302788844621514
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"current_steps": 7139,
|
| 670 |
+
"loss": 1.6836,
|
| 671 |
+
"learning_rate": 0.0007164948453608248,
|
| 672 |
+
"epoch": 2.8557768924302787
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"current_steps": 7203,
|
| 676 |
+
"loss": 1.5774,
|
| 677 |
+
"learning_rate": 0.0007139175257731959,
|
| 678 |
+
"epoch": 2.8812749003984064
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"current_steps": 7267,
|
| 682 |
+
"loss": 1.6251,
|
| 683 |
+
"learning_rate": 0.0007113402061855671,
|
| 684 |
+
"epoch": 2.906772908366534
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"current_steps": 7331,
|
| 688 |
+
"loss": 1.6695,
|
| 689 |
+
"learning_rate": 0.0007087628865979382,
|
| 690 |
+
"epoch": 2.9322709163346614
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"current_steps": 7395,
|
| 694 |
+
"loss": 1.6645,
|
| 695 |
+
"learning_rate": 0.0007061855670103094,
|
| 696 |
+
"epoch": 2.9577689243027887
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"current_steps": 7459,
|
| 700 |
+
"loss": 1.6589,
|
| 701 |
+
"learning_rate": 0.0007036082474226805,
|
| 702 |
+
"epoch": 2.9832669322709164
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"current_steps": 7509,
|
| 706 |
+
"loss": 1.4876,
|
| 707 |
+
"learning_rate": 0.0007010309278350515,
|
| 708 |
+
"epoch": 3.0087649402390437
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"current_steps": 7573,
|
| 712 |
+
"loss": 1.3509,
|
| 713 |
+
"learning_rate": 0.0006984536082474226,
|
| 714 |
+
"epoch": 3.0342629482071715
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"current_steps": 7637,
|
| 718 |
+
"loss": 1.3058,
|
| 719 |
+
"learning_rate": 0.0006958762886597938,
|
| 720 |
+
"epoch": 3.0597609561752988
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"current_steps": 7701,
|
| 724 |
+
"loss": 1.4409,
|
| 725 |
+
"learning_rate": 0.0006932989690721649,
|
| 726 |
+
"epoch": 3.0852589641434265
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"current_steps": 7765,
|
| 730 |
+
"loss": 1.3829,
|
| 731 |
+
"learning_rate": 0.0006907216494845361,
|
| 732 |
+
"epoch": 3.1107569721115538
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"current_steps": 7829,
|
| 736 |
+
"loss": 1.3513,
|
| 737 |
+
"learning_rate": 0.0006881443298969072,
|
| 738 |
+
"epoch": 3.1362549800796815
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"current_steps": 7893,
|
| 742 |
+
"loss": 1.3223,
|
| 743 |
+
"learning_rate": 0.0006855670103092784,
|
| 744 |
+
"epoch": 3.161752988047809
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"current_steps": 7957,
|
| 748 |
+
"loss": 1.2705,
|
| 749 |
+
"learning_rate": 0.0006829896907216495,
|
| 750 |
+
"epoch": 3.187250996015936
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"current_steps": 8021,
|
| 754 |
+
"loss": 1.3133,
|
| 755 |
+
"learning_rate": 0.0006804123711340207,
|
| 756 |
+
"epoch": 3.212749003984064
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"current_steps": 8085,
|
| 760 |
+
"loss": 1.3229,
|
| 761 |
+
"learning_rate": 0.0006778350515463918,
|
| 762 |
+
"epoch": 3.238247011952191
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"current_steps": 8149,
|
| 766 |
+
"loss": 1.3097,
|
| 767 |
+
"learning_rate": 0.000675257731958763,
|
| 768 |
+
"epoch": 3.263745019920319
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"current_steps": 8213,
|
| 772 |
+
"loss": 1.2961,
|
| 773 |
+
"learning_rate": 0.0006726804123711341,
|
| 774 |
+
"epoch": 3.289243027888446
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"current_steps": 8277,
|
| 778 |
+
"loss": 1.3048,
|
| 779 |
+
"learning_rate": 0.0006701030927835051,
|
| 780 |
+
"epoch": 3.314741035856574
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"current_steps": 8341,
|
| 784 |
+
"loss": 1.2909,
|
| 785 |
+
"learning_rate": 0.0006675257731958762,
|
| 786 |
+
"epoch": 3.340239043824701
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"current_steps": 8405,
|
| 790 |
+
"loss": 1.3333,
|
| 791 |
+
"learning_rate": 0.0006649484536082474,
|
| 792 |
+
"epoch": 3.365737051792829
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"current_steps": 8469,
|
| 796 |
+
"loss": 1.2552,
|
| 797 |
+
"learning_rate": 0.0006623711340206185,
|
| 798 |
+
"epoch": 3.391235059760956
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"current_steps": 8533,
|
| 802 |
+
"loss": 1.306,
|
| 803 |
+
"learning_rate": 0.0006597938144329897,
|
| 804 |
+
"epoch": 3.4167330677290835
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"current_steps": 8597,
|
| 808 |
+
"loss": 1.2382,
|
| 809 |
+
"learning_rate": 0.0006572164948453608,
|
| 810 |
+
"epoch": 3.442231075697211
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"current_steps": 8661,
|
| 814 |
+
"loss": 1.2718,
|
| 815 |
+
"learning_rate": 0.000654639175257732,
|
| 816 |
+
"epoch": 3.4677290836653385
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"current_steps": 8725,
|
| 820 |
+
"loss": 1.2348,
|
| 821 |
+
"learning_rate": 0.0006520618556701031,
|
| 822 |
+
"epoch": 3.4932270916334662
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"current_steps": 8789,
|
| 826 |
+
"loss": 1.2724,
|
| 827 |
+
"learning_rate": 0.0006494845360824743,
|
| 828 |
+
"epoch": 3.5187250996015935
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"current_steps": 8853,
|
| 832 |
+
"loss": 1.244,
|
| 833 |
+
"learning_rate": 0.0006469072164948454,
|
| 834 |
+
"epoch": 3.5442231075697213
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"current_steps": 8917,
|
| 838 |
+
"loss": 1.2948,
|
| 839 |
+
"learning_rate": 0.0006443298969072166,
|
| 840 |
+
"epoch": 3.5697211155378485
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"current_steps": 8981,
|
| 844 |
+
"loss": 1.2063,
|
| 845 |
+
"learning_rate": 0.0006417525773195877,
|
| 846 |
+
"epoch": 3.5952191235059763
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"current_steps": 9045,
|
| 850 |
+
"loss": 1.2656,
|
| 851 |
+
"learning_rate": 0.0006391752577319587,
|
| 852 |
+
"epoch": 3.6207171314741036
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"current_steps": 9109,
|
| 856 |
+
"loss": 1.25,
|
| 857 |
+
"learning_rate": 0.0006365979381443298,
|
| 858 |
+
"epoch": 3.646215139442231
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"current_steps": 9173,
|
| 862 |
+
"loss": 1.3042,
|
| 863 |
+
"learning_rate": 0.000634020618556701,
|
| 864 |
+
"epoch": 3.6717131474103586
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"current_steps": 9237,
|
| 868 |
+
"loss": 1.2612,
|
| 869 |
+
"learning_rate": 0.0006314432989690721,
|
| 870 |
+
"epoch": 3.6972111553784863
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"current_steps": 9301,
|
| 874 |
+
"loss": 1.2516,
|
| 875 |
+
"learning_rate": 0.0006288659793814433,
|
| 876 |
+
"epoch": 3.7227091633466136
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"current_steps": 9365,
|
| 880 |
+
"loss": 1.2572,
|
| 881 |
+
"learning_rate": 0.0006262886597938144,
|
| 882 |
+
"epoch": 3.748207171314741
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"current_steps": 9429,
|
| 886 |
+
"loss": 1.2525,
|
| 887 |
+
"learning_rate": 0.0006237113402061856,
|
| 888 |
+
"epoch": 3.7737051792828686
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"current_steps": 9493,
|
| 892 |
+
"loss": 1.2509,
|
| 893 |
+
"learning_rate": 0.0006211340206185567,
|
| 894 |
+
"epoch": 3.799203187250996
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"current_steps": 9557,
|
| 898 |
+
"loss": 1.2467,
|
| 899 |
+
"learning_rate": 0.0006185567010309279,
|
| 900 |
+
"epoch": 3.8247011952191237
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"current_steps": 9621,
|
| 904 |
+
"loss": 1.2375,
|
| 905 |
+
"learning_rate": 0.000615979381443299,
|
| 906 |
+
"epoch": 3.850199203187251
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"current_steps": 9685,
|
| 910 |
+
"loss": 1.2406,
|
| 911 |
+
"learning_rate": 0.0006134020618556702,
|
| 912 |
+
"epoch": 3.8756972111553782
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"current_steps": 9749,
|
| 916 |
+
"loss": 1.2517,
|
| 917 |
+
"learning_rate": 0.0006108247422680413,
|
| 918 |
+
"epoch": 3.901195219123506
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"current_steps": 9813,
|
| 922 |
+
"loss": 1.2533,
|
| 923 |
+
"learning_rate": 0.0006082474226804123,
|
| 924 |
+
"epoch": 3.9266932270916337
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"current_steps": 9877,
|
| 928 |
+
"loss": 1.256,
|
| 929 |
+
"learning_rate": 0.0006056701030927834,
|
| 930 |
+
"epoch": 3.952191235059761
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"current_steps": 9941,
|
| 934 |
+
"loss": 1.2826,
|
| 935 |
+
"learning_rate": 0.0006030927835051546,
|
| 936 |
+
"epoch": 3.9776892430278883
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"current_steps": 9991,
|
| 940 |
+
"loss": 1.226,
|
| 941 |
+
"learning_rate": 0.0006005154639175257,
|
| 942 |
+
"epoch": 4.003187250996016
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"current_steps": 10055,
|
| 946 |
+
"loss": 0.9512,
|
| 947 |
+
"learning_rate": 0.0005979381443298969,
|
| 948 |
+
"epoch": 4.028685258964144
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"current_steps": 10055,
|
| 952 |
+
"loss": 0.9512,
|
| 953 |
+
"learning_rate": 0.0005979381443298969,
|
| 954 |
+
"epoch": 4.028685258964144
|
| 955 |
+
}
|
| 956 |
+
]
|
training_graph.png
ADDED
|
training_log.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"base_model_name": "meta-llama_Meta-Llama-3-8B",
|
| 3 |
+
"base_model_class": "LlamaForCausalLM",
|
| 4 |
+
"base_loaded_in_4bit": false,
|
| 5 |
+
"base_loaded_in_8bit": false,
|
| 6 |
+
"projections": "q, v",
|
| 7 |
+
"loss": 0.9512,
|
| 8 |
+
"grad_norm": 0.6213752031326294,
|
| 9 |
+
"learning_rate": 0.0005979381443298969,
|
| 10 |
+
"epoch": 4.028685258964144,
|
| 11 |
+
"current_steps": 10055,
|
| 12 |
+
"current_steps_adjusted": 10055,
|
| 13 |
+
"epoch_adjusted": 4.028685258964144,
|
| 14 |
+
"train_runtime": 3864.6216,
|
| 15 |
+
"train_samples_per_second": 25.977,
|
| 16 |
+
"train_steps_per_second": 0.101,
|
| 17 |
+
"total_flos": 4.69608070712918e+17,
|
| 18 |
+
"train_loss": 1.8759732170950008
|
| 19 |
+
}
|
training_parameters.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"lora_name": "lora-meta-llama_Meta-Llama-3-8B-16bit-ORKL-cleaned_juergen-v3_LR_1e-3",
|
| 3 |
+
"always_override": true,
|
| 4 |
+
"save_steps": 722,
|
| 5 |
+
"micro_batch_size": 4,
|
| 6 |
+
"batch_size": 0,
|
| 7 |
+
"epochs": 10,
|
| 8 |
+
"learning_rate": "1e-3",
|
| 9 |
+
"lr_scheduler_type": "linear",
|
| 10 |
+
"lora_rank": 128,
|
| 11 |
+
"lora_alpha": 256,
|
| 12 |
+
"lora_dropout": 0.05,
|
| 13 |
+
"cutoff_len": 256,
|
| 14 |
+
"dataset": "None",
|
| 15 |
+
"eval_dataset": "None",
|
| 16 |
+
"format": "None",
|
| 17 |
+
"eval_steps": 100,
|
| 18 |
+
"raw_text_file": "orkl-cleaned-juergen",
|
| 19 |
+
"higher_rank_limit": false,
|
| 20 |
+
"warmup_steps": 128,
|
| 21 |
+
"optimizer": "adamw_torch",
|
| 22 |
+
"hard_cut_string": "\\n\\n\\n",
|
| 23 |
+
"train_only_after": "",
|
| 24 |
+
"stop_at_loss": 1,
|
| 25 |
+
"add_eos_token": false,
|
| 26 |
+
"min_chars": 3,
|
| 27 |
+
"report_to": "wandb",
|
| 28 |
+
"precize_slicing_overlap": true,
|
| 29 |
+
"add_eos_token_type": "Every Block",
|
| 30 |
+
"save_steps_under_loss": 1.8,
|
| 31 |
+
"add_bos_token": true,
|
| 32 |
+
"training_projection": "q-v",
|
| 33 |
+
"sliding_window": false,
|
| 34 |
+
"warmup_ratio": 0,
|
| 35 |
+
"grad_accumulation": 64,
|
| 36 |
+
"neft_noise_alpha": 0
|
| 37 |
+
}
|
training_prompt.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"template_type": "raw_text"
|
| 3 |
+
}
|