Upload folder using huggingface_hub
Browse files- README.md +3 -3
- adapter_config.json +35 -0
- adapter_model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1433 -0
- training_args.bin +3 -0
README.md
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
---
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
---
|
adapter_config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "/root/autodl-tmp/Baichuan-M1-14B-Instruct",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"eva_config": null,
|
| 7 |
+
"exclude_modules": null,
|
| 8 |
+
"fan_in_fan_out": false,
|
| 9 |
+
"inference_mode": true,
|
| 10 |
+
"init_lora_weights": true,
|
| 11 |
+
"layer_replication": null,
|
| 12 |
+
"layers_pattern": null,
|
| 13 |
+
"layers_to_transform": null,
|
| 14 |
+
"loftq_config": {},
|
| 15 |
+
"lora_alpha": 32,
|
| 16 |
+
"lora_bias": false,
|
| 17 |
+
"lora_dropout": 0.1,
|
| 18 |
+
"megatron_config": null,
|
| 19 |
+
"megatron_core": "megatron.core",
|
| 20 |
+
"modules_to_save": null,
|
| 21 |
+
"peft_type": "LORA",
|
| 22 |
+
"r": 16,
|
| 23 |
+
"rank_pattern": {},
|
| 24 |
+
"revision": null,
|
| 25 |
+
"target_modules": [
|
| 26 |
+
"W_pack",
|
| 27 |
+
"gate_proj",
|
| 28 |
+
"up_proj",
|
| 29 |
+
"down_proj",
|
| 30 |
+
"o_proj"
|
| 31 |
+
],
|
| 32 |
+
"task_type": "CAUSAL_LM",
|
| 33 |
+
"use_dora": false,
|
| 34 |
+
"use_rslora": false
|
| 35 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a238c0e1343f1a6b62af2c467f843e3eae4af0cbb239ba4e7145297ed5d7738
|
| 3 |
+
size 229429600
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3917a15c0d17b3e270f79d827c2738ac9350c87f8edbc6ee0e420a70fa1574e2
|
| 3 |
+
size 459089602
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a15993cad901dc5280fe73952d778614791cf68373eadc2f1a256a37bca2505f
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd0991e79159ea97a3cbedcd65d3728a0a322a29100cc8b6d96d2321a28da7fb
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1433 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.07272727272727272,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0003636363636363636,
|
| 13 |
+
"grad_norm": 0.47871133685112,
|
| 14 |
+
"learning_rate": 4.999090909090909e-05,
|
| 15 |
+
"loss": 1.4412,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0007272727272727272,
|
| 20 |
+
"grad_norm": 0.4764420688152313,
|
| 21 |
+
"learning_rate": 4.998181818181818e-05,
|
| 22 |
+
"loss": 1.3082,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.001090909090909091,
|
| 27 |
+
"grad_norm": 0.6560420989990234,
|
| 28 |
+
"learning_rate": 4.997272727272728e-05,
|
| 29 |
+
"loss": 1.3592,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0014545454545454545,
|
| 34 |
+
"grad_norm": 0.7296062111854553,
|
| 35 |
+
"learning_rate": 4.996363636363637e-05,
|
| 36 |
+
"loss": 1.1219,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0018181818181818182,
|
| 41 |
+
"grad_norm": 0.46472349762916565,
|
| 42 |
+
"learning_rate": 4.995454545454546e-05,
|
| 43 |
+
"loss": 1.0707,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.002181818181818182,
|
| 48 |
+
"grad_norm": 0.5377005338668823,
|
| 49 |
+
"learning_rate": 4.994545454545454e-05,
|
| 50 |
+
"loss": 1.1453,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0025454545454545456,
|
| 55 |
+
"grad_norm": 0.5077579021453857,
|
| 56 |
+
"learning_rate": 4.993636363636364e-05,
|
| 57 |
+
"loss": 1.1129,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.002909090909090909,
|
| 62 |
+
"grad_norm": 0.5722835063934326,
|
| 63 |
+
"learning_rate": 4.992727272727273e-05,
|
| 64 |
+
"loss": 1.1029,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0032727272727272726,
|
| 69 |
+
"grad_norm": 0.5486398339271545,
|
| 70 |
+
"learning_rate": 4.991818181818182e-05,
|
| 71 |
+
"loss": 1.0021,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0036363636363636364,
|
| 76 |
+
"grad_norm": 0.6710432171821594,
|
| 77 |
+
"learning_rate": 4.990909090909091e-05,
|
| 78 |
+
"loss": 1.0766,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.004,
|
| 83 |
+
"grad_norm": 0.45256921648979187,
|
| 84 |
+
"learning_rate": 4.99e-05,
|
| 85 |
+
"loss": 1.0424,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.004363636363636364,
|
| 90 |
+
"grad_norm": 0.5884903073310852,
|
| 91 |
+
"learning_rate": 4.98909090909091e-05,
|
| 92 |
+
"loss": 1.1271,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0047272727272727275,
|
| 97 |
+
"grad_norm": 0.8489357233047485,
|
| 98 |
+
"learning_rate": 4.988181818181819e-05,
|
| 99 |
+
"loss": 1.1285,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.005090909090909091,
|
| 104 |
+
"grad_norm": 0.6435407996177673,
|
| 105 |
+
"learning_rate": 4.987272727272728e-05,
|
| 106 |
+
"loss": 1.1967,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.005454545454545455,
|
| 111 |
+
"grad_norm": 0.5722768306732178,
|
| 112 |
+
"learning_rate": 4.986363636363637e-05,
|
| 113 |
+
"loss": 1.0342,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.005818181818181818,
|
| 118 |
+
"grad_norm": 0.4862392544746399,
|
| 119 |
+
"learning_rate": 4.985454545454546e-05,
|
| 120 |
+
"loss": 0.9976,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.0061818181818181816,
|
| 125 |
+
"grad_norm": 0.7204858064651489,
|
| 126 |
+
"learning_rate": 4.984545454545455e-05,
|
| 127 |
+
"loss": 1.0818,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.006545454545454545,
|
| 132 |
+
"grad_norm": 0.727836549282074,
|
| 133 |
+
"learning_rate": 4.983636363636364e-05,
|
| 134 |
+
"loss": 1.0893,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.006909090909090909,
|
| 139 |
+
"grad_norm": 0.4965401589870453,
|
| 140 |
+
"learning_rate": 4.982727272727273e-05,
|
| 141 |
+
"loss": 1.01,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.007272727272727273,
|
| 146 |
+
"grad_norm": 0.6770114898681641,
|
| 147 |
+
"learning_rate": 4.981818181818182e-05,
|
| 148 |
+
"loss": 0.9949,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.0076363636363636364,
|
| 153 |
+
"grad_norm": 0.795165479183197,
|
| 154 |
+
"learning_rate": 4.980909090909091e-05,
|
| 155 |
+
"loss": 1.1506,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.008,
|
| 160 |
+
"grad_norm": 0.7894571423530579,
|
| 161 |
+
"learning_rate": 4.9800000000000004e-05,
|
| 162 |
+
"loss": 1.0309,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.008363636363636363,
|
| 167 |
+
"grad_norm": 0.9198132753372192,
|
| 168 |
+
"learning_rate": 4.9790909090909094e-05,
|
| 169 |
+
"loss": 0.9611,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.008727272727272728,
|
| 174 |
+
"grad_norm": 0.8672171831130981,
|
| 175 |
+
"learning_rate": 4.9781818181818184e-05,
|
| 176 |
+
"loss": 1.1689,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.00909090909090909,
|
| 181 |
+
"grad_norm": 0.6119312644004822,
|
| 182 |
+
"learning_rate": 4.9772727272727275e-05,
|
| 183 |
+
"loss": 0.9891,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.009454545454545455,
|
| 188 |
+
"grad_norm": 0.7098168134689331,
|
| 189 |
+
"learning_rate": 4.9763636363636365e-05,
|
| 190 |
+
"loss": 0.9369,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.009818181818181818,
|
| 195 |
+
"grad_norm": 0.7419410347938538,
|
| 196 |
+
"learning_rate": 4.975454545454546e-05,
|
| 197 |
+
"loss": 1.0293,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.010181818181818183,
|
| 202 |
+
"grad_norm": 0.6659941673278809,
|
| 203 |
+
"learning_rate": 4.9745454545454545e-05,
|
| 204 |
+
"loss": 1.0936,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.010545454545454545,
|
| 209 |
+
"grad_norm": 0.8974291682243347,
|
| 210 |
+
"learning_rate": 4.9736363636363635e-05,
|
| 211 |
+
"loss": 1.1088,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.01090909090909091,
|
| 216 |
+
"grad_norm": 0.7319221496582031,
|
| 217 |
+
"learning_rate": 4.9727272727272725e-05,
|
| 218 |
+
"loss": 0.9916,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.011272727272727273,
|
| 223 |
+
"grad_norm": 0.9040235280990601,
|
| 224 |
+
"learning_rate": 4.971818181818182e-05,
|
| 225 |
+
"loss": 1.1141,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.011636363636363636,
|
| 230 |
+
"grad_norm": 0.7666190266609192,
|
| 231 |
+
"learning_rate": 4.970909090909091e-05,
|
| 232 |
+
"loss": 1.0309,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.012,
|
| 237 |
+
"grad_norm": 0.7975471615791321,
|
| 238 |
+
"learning_rate": 4.97e-05,
|
| 239 |
+
"loss": 1.0,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.012363636363636363,
|
| 244 |
+
"grad_norm": 0.7558386325836182,
|
| 245 |
+
"learning_rate": 4.969090909090909e-05,
|
| 246 |
+
"loss": 1.1283,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.012727272727272728,
|
| 251 |
+
"grad_norm": 0.76316899061203,
|
| 252 |
+
"learning_rate": 4.968181818181818e-05,
|
| 253 |
+
"loss": 1.01,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.01309090909090909,
|
| 258 |
+
"grad_norm": 0.5065900683403015,
|
| 259 |
+
"learning_rate": 4.967272727272728e-05,
|
| 260 |
+
"loss": 1.002,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.013454545454545455,
|
| 265 |
+
"grad_norm": 1.0256422758102417,
|
| 266 |
+
"learning_rate": 4.966363636363637e-05,
|
| 267 |
+
"loss": 1.0912,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.013818181818181818,
|
| 272 |
+
"grad_norm": 0.734935998916626,
|
| 273 |
+
"learning_rate": 4.965454545454546e-05,
|
| 274 |
+
"loss": 1.102,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.014181818181818183,
|
| 279 |
+
"grad_norm": 0.6911085844039917,
|
| 280 |
+
"learning_rate": 4.964545454545455e-05,
|
| 281 |
+
"loss": 1.0479,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.014545454545454545,
|
| 286 |
+
"grad_norm": 0.778714120388031,
|
| 287 |
+
"learning_rate": 4.963636363636364e-05,
|
| 288 |
+
"loss": 1.0379,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.014909090909090908,
|
| 293 |
+
"grad_norm": 0.6882097721099854,
|
| 294 |
+
"learning_rate": 4.962727272727273e-05,
|
| 295 |
+
"loss": 0.9732,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.015272727272727273,
|
| 300 |
+
"grad_norm": 0.747656524181366,
|
| 301 |
+
"learning_rate": 4.961818181818182e-05,
|
| 302 |
+
"loss": 1.0602,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.015636363636363636,
|
| 307 |
+
"grad_norm": 0.7968394160270691,
|
| 308 |
+
"learning_rate": 4.960909090909091e-05,
|
| 309 |
+
"loss": 1.1314,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.016,
|
| 314 |
+
"grad_norm": 0.9477614760398865,
|
| 315 |
+
"learning_rate": 4.96e-05,
|
| 316 |
+
"loss": 1.0488,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.016363636363636365,
|
| 321 |
+
"grad_norm": 0.7730783820152283,
|
| 322 |
+
"learning_rate": 4.9590909090909096e-05,
|
| 323 |
+
"loss": 1.1314,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.016727272727272726,
|
| 328 |
+
"grad_norm": 1.1297029256820679,
|
| 329 |
+
"learning_rate": 4.9581818181818186e-05,
|
| 330 |
+
"loss": 1.0943,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.01709090909090909,
|
| 335 |
+
"grad_norm": 0.7387447357177734,
|
| 336 |
+
"learning_rate": 4.9572727272727276e-05,
|
| 337 |
+
"loss": 1.0711,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.017454545454545455,
|
| 342 |
+
"grad_norm": 0.5228479504585266,
|
| 343 |
+
"learning_rate": 4.9563636363636367e-05,
|
| 344 |
+
"loss": 1.0711,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.01781818181818182,
|
| 349 |
+
"grad_norm": 0.5137602090835571,
|
| 350 |
+
"learning_rate": 4.9554545454545457e-05,
|
| 351 |
+
"loss": 0.9242,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.01818181818181818,
|
| 356 |
+
"grad_norm": 0.859491229057312,
|
| 357 |
+
"learning_rate": 4.9545454545454553e-05,
|
| 358 |
+
"loss": 1.124,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.018545454545454546,
|
| 363 |
+
"grad_norm": 0.7495261430740356,
|
| 364 |
+
"learning_rate": 4.953636363636364e-05,
|
| 365 |
+
"loss": 1.1184,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.01890909090909091,
|
| 370 |
+
"grad_norm": 0.8758722543716431,
|
| 371 |
+
"learning_rate": 4.952727272727273e-05,
|
| 372 |
+
"loss": 1.1215,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.01927272727272727,
|
| 377 |
+
"grad_norm": 0.7047644257545471,
|
| 378 |
+
"learning_rate": 4.951818181818182e-05,
|
| 379 |
+
"loss": 1.0416,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.019636363636363636,
|
| 384 |
+
"grad_norm": 0.9179386496543884,
|
| 385 |
+
"learning_rate": 4.9509090909090914e-05,
|
| 386 |
+
"loss": 1.0404,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.02,
|
| 391 |
+
"grad_norm": 0.8945435881614685,
|
| 392 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 393 |
+
"loss": 1.0211,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.020363636363636365,
|
| 398 |
+
"grad_norm": 0.49766021966934204,
|
| 399 |
+
"learning_rate": 4.9490909090909094e-05,
|
| 400 |
+
"loss": 0.9262,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.020727272727272726,
|
| 405 |
+
"grad_norm": 0.8015190958976746,
|
| 406 |
+
"learning_rate": 4.9481818181818184e-05,
|
| 407 |
+
"loss": 1.085,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.02109090909090909,
|
| 412 |
+
"grad_norm": 0.8135480284690857,
|
| 413 |
+
"learning_rate": 4.9472727272727274e-05,
|
| 414 |
+
"loss": 0.9673,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.021454545454545455,
|
| 419 |
+
"grad_norm": 0.7892040610313416,
|
| 420 |
+
"learning_rate": 4.946363636363637e-05,
|
| 421 |
+
"loss": 1.0301,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.02181818181818182,
|
| 426 |
+
"grad_norm": 0.6644595265388489,
|
| 427 |
+
"learning_rate": 4.945454545454546e-05,
|
| 428 |
+
"loss": 0.9969,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.02218181818181818,
|
| 433 |
+
"grad_norm": 0.6504725813865662,
|
| 434 |
+
"learning_rate": 4.944545454545455e-05,
|
| 435 |
+
"loss": 1.0686,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.022545454545454546,
|
| 440 |
+
"grad_norm": 0.5474579930305481,
|
| 441 |
+
"learning_rate": 4.9436363636363634e-05,
|
| 442 |
+
"loss": 1.0338,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.02290909090909091,
|
| 447 |
+
"grad_norm": 1.133872151374817,
|
| 448 |
+
"learning_rate": 4.9427272727272724e-05,
|
| 449 |
+
"loss": 1.0355,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.02327272727272727,
|
| 454 |
+
"grad_norm": 0.9010602235794067,
|
| 455 |
+
"learning_rate": 4.941818181818182e-05,
|
| 456 |
+
"loss": 1.0205,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.023636363636363636,
|
| 461 |
+
"grad_norm": 0.6271716952323914,
|
| 462 |
+
"learning_rate": 4.940909090909091e-05,
|
| 463 |
+
"loss": 0.9688,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.024,
|
| 468 |
+
"grad_norm": 0.7340686917304993,
|
| 469 |
+
"learning_rate": 4.94e-05,
|
| 470 |
+
"loss": 1.058,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.024363636363636365,
|
| 475 |
+
"grad_norm": 0.9811447262763977,
|
| 476 |
+
"learning_rate": 4.939090909090909e-05,
|
| 477 |
+
"loss": 1.0941,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.024727272727272726,
|
| 482 |
+
"grad_norm": 0.8026047348976135,
|
| 483 |
+
"learning_rate": 4.938181818181818e-05,
|
| 484 |
+
"loss": 1.0148,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.02509090909090909,
|
| 489 |
+
"grad_norm": 0.8320631980895996,
|
| 490 |
+
"learning_rate": 4.937272727272728e-05,
|
| 491 |
+
"loss": 1.0332,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.025454545454545455,
|
| 496 |
+
"grad_norm": 1.0077319145202637,
|
| 497 |
+
"learning_rate": 4.936363636363637e-05,
|
| 498 |
+
"loss": 0.882,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.025818181818181817,
|
| 503 |
+
"grad_norm": 0.792972981929779,
|
| 504 |
+
"learning_rate": 4.935454545454546e-05,
|
| 505 |
+
"loss": 0.9613,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.02618181818181818,
|
| 510 |
+
"grad_norm": 0.9640710353851318,
|
| 511 |
+
"learning_rate": 4.934545454545455e-05,
|
| 512 |
+
"loss": 0.9879,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.026545454545454546,
|
| 517 |
+
"grad_norm": 0.6762244701385498,
|
| 518 |
+
"learning_rate": 4.933636363636364e-05,
|
| 519 |
+
"loss": 0.9643,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.02690909090909091,
|
| 524 |
+
"grad_norm": 0.8114360570907593,
|
| 525 |
+
"learning_rate": 4.932727272727273e-05,
|
| 526 |
+
"loss": 1.0049,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.02727272727272727,
|
| 531 |
+
"grad_norm": 0.6531985998153687,
|
| 532 |
+
"learning_rate": 4.931818181818182e-05,
|
| 533 |
+
"loss": 0.9141,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.027636363636363636,
|
| 538 |
+
"grad_norm": 0.846741795539856,
|
| 539 |
+
"learning_rate": 4.930909090909091e-05,
|
| 540 |
+
"loss": 0.9738,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.028,
|
| 545 |
+
"grad_norm": 0.8146620392799377,
|
| 546 |
+
"learning_rate": 4.93e-05,
|
| 547 |
+
"loss": 1.0508,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.028363636363636365,
|
| 552 |
+
"grad_norm": 0.8905590176582336,
|
| 553 |
+
"learning_rate": 4.9290909090909096e-05,
|
| 554 |
+
"loss": 0.9883,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.028727272727272726,
|
| 559 |
+
"grad_norm": 0.820004403591156,
|
| 560 |
+
"learning_rate": 4.9281818181818186e-05,
|
| 561 |
+
"loss": 1.1111,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.02909090909090909,
|
| 566 |
+
"grad_norm": 0.6964171528816223,
|
| 567 |
+
"learning_rate": 4.9272727272727276e-05,
|
| 568 |
+
"loss": 0.9889,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.029454545454545455,
|
| 573 |
+
"grad_norm": 0.7463963627815247,
|
| 574 |
+
"learning_rate": 4.9263636363636366e-05,
|
| 575 |
+
"loss": 1.0017,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.029818181818181817,
|
| 580 |
+
"grad_norm": 1.1425820589065552,
|
| 581 |
+
"learning_rate": 4.9254545454545456e-05,
|
| 582 |
+
"loss": 1.0291,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.03018181818181818,
|
| 587 |
+
"grad_norm": 0.8121577501296997,
|
| 588 |
+
"learning_rate": 4.924545454545455e-05,
|
| 589 |
+
"loss": 1.1896,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.030545454545454546,
|
| 594 |
+
"grad_norm": 0.7760348320007324,
|
| 595 |
+
"learning_rate": 4.9236363636363636e-05,
|
| 596 |
+
"loss": 1.0086,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.03090909090909091,
|
| 601 |
+
"grad_norm": 1.0461779832839966,
|
| 602 |
+
"learning_rate": 4.9227272727272726e-05,
|
| 603 |
+
"loss": 1.067,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.03127272727272727,
|
| 608 |
+
"grad_norm": 0.7449011206626892,
|
| 609 |
+
"learning_rate": 4.9218181818181816e-05,
|
| 610 |
+
"loss": 0.9166,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.031636363636363636,
|
| 615 |
+
"grad_norm": 0.8614558577537537,
|
| 616 |
+
"learning_rate": 4.920909090909091e-05,
|
| 617 |
+
"loss": 1.0984,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.032,
|
| 622 |
+
"grad_norm": 0.9124776721000671,
|
| 623 |
+
"learning_rate": 4.92e-05,
|
| 624 |
+
"loss": 0.9428,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.032363636363636365,
|
| 629 |
+
"grad_norm": 0.7346855998039246,
|
| 630 |
+
"learning_rate": 4.919090909090909e-05,
|
| 631 |
+
"loss": 0.9891,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.03272727272727273,
|
| 636 |
+
"grad_norm": 0.6888474822044373,
|
| 637 |
+
"learning_rate": 4.9181818181818183e-05,
|
| 638 |
+
"loss": 1.1186,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.03309090909090909,
|
| 643 |
+
"grad_norm": 0.5666208267211914,
|
| 644 |
+
"learning_rate": 4.9172727272727273e-05,
|
| 645 |
+
"loss": 0.9862,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.03345454545454545,
|
| 650 |
+
"grad_norm": 0.6969447731971741,
|
| 651 |
+
"learning_rate": 4.916363636363637e-05,
|
| 652 |
+
"loss": 1.09,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.03381818181818182,
|
| 657 |
+
"grad_norm": 0.7007094025611877,
|
| 658 |
+
"learning_rate": 4.915454545454546e-05,
|
| 659 |
+
"loss": 0.9951,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.03418181818181818,
|
| 664 |
+
"grad_norm": 0.8678598999977112,
|
| 665 |
+
"learning_rate": 4.914545454545455e-05,
|
| 666 |
+
"loss": 0.9574,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.034545454545454546,
|
| 671 |
+
"grad_norm": 0.6946694850921631,
|
| 672 |
+
"learning_rate": 4.913636363636364e-05,
|
| 673 |
+
"loss": 1.0867,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.03490909090909091,
|
| 678 |
+
"grad_norm": 0.7773414254188538,
|
| 679 |
+
"learning_rate": 4.912727272727273e-05,
|
| 680 |
+
"loss": 1.0025,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.035272727272727275,
|
| 685 |
+
"grad_norm": 0.8114984631538391,
|
| 686 |
+
"learning_rate": 4.911818181818182e-05,
|
| 687 |
+
"loss": 0.9855,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.03563636363636364,
|
| 692 |
+
"grad_norm": 0.6157673597335815,
|
| 693 |
+
"learning_rate": 4.910909090909091e-05,
|
| 694 |
+
"loss": 0.9988,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.036,
|
| 699 |
+
"grad_norm": 0.8051506280899048,
|
| 700 |
+
"learning_rate": 4.91e-05,
|
| 701 |
+
"loss": 0.9818,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.03636363636363636,
|
| 706 |
+
"grad_norm": 0.8241102695465088,
|
| 707 |
+
"learning_rate": 4.909090909090909e-05,
|
| 708 |
+
"loss": 0.9893,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.036727272727272726,
|
| 713 |
+
"grad_norm": 0.7396079897880554,
|
| 714 |
+
"learning_rate": 4.908181818181818e-05,
|
| 715 |
+
"loss": 0.9775,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.03709090909090909,
|
| 720 |
+
"grad_norm": 0.730338454246521,
|
| 721 |
+
"learning_rate": 4.907272727272728e-05,
|
| 722 |
+
"loss": 1.0172,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.037454545454545456,
|
| 727 |
+
"grad_norm": 0.9117375016212463,
|
| 728 |
+
"learning_rate": 4.906363636363637e-05,
|
| 729 |
+
"loss": 1.0273,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.03781818181818182,
|
| 734 |
+
"grad_norm": 0.699874997138977,
|
| 735 |
+
"learning_rate": 4.905454545454546e-05,
|
| 736 |
+
"loss": 0.9398,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.038181818181818185,
|
| 741 |
+
"grad_norm": 0.724445641040802,
|
| 742 |
+
"learning_rate": 4.904545454545455e-05,
|
| 743 |
+
"loss": 0.9617,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.03854545454545454,
|
| 748 |
+
"grad_norm": 0.7848714590072632,
|
| 749 |
+
"learning_rate": 4.903636363636364e-05,
|
| 750 |
+
"loss": 1.0037,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.03890909090909091,
|
| 755 |
+
"grad_norm": 1.0549767017364502,
|
| 756 |
+
"learning_rate": 4.902727272727273e-05,
|
| 757 |
+
"loss": 1.0691,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.03927272727272727,
|
| 762 |
+
"grad_norm": 0.8227099776268005,
|
| 763 |
+
"learning_rate": 4.901818181818182e-05,
|
| 764 |
+
"loss": 1.0037,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.039636363636363636,
|
| 769 |
+
"grad_norm": 1.0453096628189087,
|
| 770 |
+
"learning_rate": 4.900909090909091e-05,
|
| 771 |
+
"loss": 1.0145,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.04,
|
| 776 |
+
"grad_norm": 0.8218994140625,
|
| 777 |
+
"learning_rate": 4.9e-05,
|
| 778 |
+
"loss": 0.9857,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.040363636363636365,
|
| 783 |
+
"grad_norm": 0.76604825258255,
|
| 784 |
+
"learning_rate": 4.8990909090909095e-05,
|
| 785 |
+
"loss": 0.9576,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.04072727272727273,
|
| 790 |
+
"grad_norm": 0.7639636397361755,
|
| 791 |
+
"learning_rate": 4.8981818181818185e-05,
|
| 792 |
+
"loss": 1.0402,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.04109090909090909,
|
| 797 |
+
"grad_norm": 0.8489947319030762,
|
| 798 |
+
"learning_rate": 4.8972727272727275e-05,
|
| 799 |
+
"loss": 1.023,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.04145454545454545,
|
| 804 |
+
"grad_norm": 0.7406574487686157,
|
| 805 |
+
"learning_rate": 4.8963636363636365e-05,
|
| 806 |
+
"loss": 1.0324,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.04181818181818182,
|
| 811 |
+
"grad_norm": 0.6394333243370056,
|
| 812 |
+
"learning_rate": 4.8954545454545456e-05,
|
| 813 |
+
"loss": 0.9831,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.04218181818181818,
|
| 818 |
+
"grad_norm": 0.9677541851997375,
|
| 819 |
+
"learning_rate": 4.894545454545455e-05,
|
| 820 |
+
"loss": 0.9928,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.042545454545454546,
|
| 825 |
+
"grad_norm": 0.7677833437919617,
|
| 826 |
+
"learning_rate": 4.893636363636364e-05,
|
| 827 |
+
"loss": 0.9301,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.04290909090909091,
|
| 832 |
+
"grad_norm": 0.6206602454185486,
|
| 833 |
+
"learning_rate": 4.8927272727272726e-05,
|
| 834 |
+
"loss": 1.0012,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.043272727272727275,
|
| 839 |
+
"grad_norm": 0.8425969481468201,
|
| 840 |
+
"learning_rate": 4.8918181818181816e-05,
|
| 841 |
+
"loss": 1.1275,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.04363636363636364,
|
| 846 |
+
"grad_norm": 0.7413673400878906,
|
| 847 |
+
"learning_rate": 4.890909090909091e-05,
|
| 848 |
+
"loss": 1.0117,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.044,
|
| 853 |
+
"grad_norm": 0.9616818428039551,
|
| 854 |
+
"learning_rate": 4.89e-05,
|
| 855 |
+
"loss": 0.9658,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.04436363636363636,
|
| 860 |
+
"grad_norm": 0.7987754344940186,
|
| 861 |
+
"learning_rate": 4.889090909090909e-05,
|
| 862 |
+
"loss": 0.9262,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.04472727272727273,
|
| 867 |
+
"grad_norm": 0.8735281825065613,
|
| 868 |
+
"learning_rate": 4.888181818181818e-05,
|
| 869 |
+
"loss": 0.9414,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.04509090909090909,
|
| 874 |
+
"grad_norm": 0.7942488789558411,
|
| 875 |
+
"learning_rate": 4.887272727272727e-05,
|
| 876 |
+
"loss": 1.0506,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.045454545454545456,
|
| 881 |
+
"grad_norm": 0.8182056546211243,
|
| 882 |
+
"learning_rate": 4.886363636363637e-05,
|
| 883 |
+
"loss": 1.0721,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.04581818181818182,
|
| 888 |
+
"grad_norm": 0.7209412455558777,
|
| 889 |
+
"learning_rate": 4.885454545454546e-05,
|
| 890 |
+
"loss": 1.0205,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.046181818181818185,
|
| 895 |
+
"grad_norm": 0.824597954750061,
|
| 896 |
+
"learning_rate": 4.884545454545455e-05,
|
| 897 |
+
"loss": 0.9563,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.04654545454545454,
|
| 902 |
+
"grad_norm": 0.6784900426864624,
|
| 903 |
+
"learning_rate": 4.883636363636364e-05,
|
| 904 |
+
"loss": 1.0971,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.04690909090909091,
|
| 909 |
+
"grad_norm": 0.680962324142456,
|
| 910 |
+
"learning_rate": 4.882727272727273e-05,
|
| 911 |
+
"loss": 1.0043,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.04727272727272727,
|
| 916 |
+
"grad_norm": 0.7917009592056274,
|
| 917 |
+
"learning_rate": 4.881818181818182e-05,
|
| 918 |
+
"loss": 1.053,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.047636363636363636,
|
| 923 |
+
"grad_norm": 0.8048487305641174,
|
| 924 |
+
"learning_rate": 4.880909090909091e-05,
|
| 925 |
+
"loss": 1.0842,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.048,
|
| 930 |
+
"grad_norm": 0.7836024165153503,
|
| 931 |
+
"learning_rate": 4.88e-05,
|
| 932 |
+
"loss": 1.0537,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.048363636363636366,
|
| 937 |
+
"grad_norm": 0.6501603126525879,
|
| 938 |
+
"learning_rate": 4.879090909090909e-05,
|
| 939 |
+
"loss": 1.0084,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.04872727272727273,
|
| 944 |
+
"grad_norm": 1.0370051860809326,
|
| 945 |
+
"learning_rate": 4.878181818181819e-05,
|
| 946 |
+
"loss": 0.9449,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.04909090909090909,
|
| 951 |
+
"grad_norm": 0.8803650736808777,
|
| 952 |
+
"learning_rate": 4.877272727272728e-05,
|
| 953 |
+
"loss": 1.0154,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.04945454545454545,
|
| 958 |
+
"grad_norm": 0.8375502824783325,
|
| 959 |
+
"learning_rate": 4.876363636363637e-05,
|
| 960 |
+
"loss": 0.9437,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.04981818181818182,
|
| 965 |
+
"grad_norm": 0.8356055617332458,
|
| 966 |
+
"learning_rate": 4.875454545454546e-05,
|
| 967 |
+
"loss": 0.9443,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.05018181818181818,
|
| 972 |
+
"grad_norm": 0.9145257472991943,
|
| 973 |
+
"learning_rate": 4.874545454545455e-05,
|
| 974 |
+
"loss": 1.0326,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.050545454545454546,
|
| 979 |
+
"grad_norm": 0.7850838899612427,
|
| 980 |
+
"learning_rate": 4.8736363636363644e-05,
|
| 981 |
+
"loss": 0.9943,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.05090909090909091,
|
| 986 |
+
"grad_norm": 0.8555696606636047,
|
| 987 |
+
"learning_rate": 4.872727272727273e-05,
|
| 988 |
+
"loss": 1.0791,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.051272727272727275,
|
| 993 |
+
"grad_norm": 0.6297926306724548,
|
| 994 |
+
"learning_rate": 4.871818181818182e-05,
|
| 995 |
+
"loss": 0.9766,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.05163636363636363,
|
| 1000 |
+
"grad_norm": 0.7590431571006775,
|
| 1001 |
+
"learning_rate": 4.870909090909091e-05,
|
| 1002 |
+
"loss": 1.0184,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.052,
|
| 1007 |
+
"grad_norm": 0.7111615538597107,
|
| 1008 |
+
"learning_rate": 4.87e-05,
|
| 1009 |
+
"loss": 0.9545,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.05236363636363636,
|
| 1014 |
+
"grad_norm": 0.9266930222511292,
|
| 1015 |
+
"learning_rate": 4.8690909090909095e-05,
|
| 1016 |
+
"loss": 0.9373,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.05272727272727273,
|
| 1021 |
+
"grad_norm": 0.7775997519493103,
|
| 1022 |
+
"learning_rate": 4.8681818181818185e-05,
|
| 1023 |
+
"loss": 1.084,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.05309090909090909,
|
| 1028 |
+
"grad_norm": 0.7686333656311035,
|
| 1029 |
+
"learning_rate": 4.8672727272727275e-05,
|
| 1030 |
+
"loss": 0.9543,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.053454545454545456,
|
| 1035 |
+
"grad_norm": 0.8209460973739624,
|
| 1036 |
+
"learning_rate": 4.8663636363636365e-05,
|
| 1037 |
+
"loss": 1.0379,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.05381818181818182,
|
| 1042 |
+
"grad_norm": 0.5993338227272034,
|
| 1043 |
+
"learning_rate": 4.8654545454545455e-05,
|
| 1044 |
+
"loss": 0.9836,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.054181818181818185,
|
| 1049 |
+
"grad_norm": 0.8321117758750916,
|
| 1050 |
+
"learning_rate": 4.864545454545455e-05,
|
| 1051 |
+
"loss": 0.9789,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.05454545454545454,
|
| 1056 |
+
"grad_norm": 0.7136039733886719,
|
| 1057 |
+
"learning_rate": 4.863636363636364e-05,
|
| 1058 |
+
"loss": 0.941,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.05490909090909091,
|
| 1063 |
+
"grad_norm": 0.6142838001251221,
|
| 1064 |
+
"learning_rate": 4.8627272727272725e-05,
|
| 1065 |
+
"loss": 1.0277,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.05527272727272727,
|
| 1070 |
+
"grad_norm": 0.7588635087013245,
|
| 1071 |
+
"learning_rate": 4.8618181818181815e-05,
|
| 1072 |
+
"loss": 1.0854,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.05563636363636364,
|
| 1077 |
+
"grad_norm": 0.9304160475730896,
|
| 1078 |
+
"learning_rate": 4.860909090909091e-05,
|
| 1079 |
+
"loss": 1.0713,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.056,
|
| 1084 |
+
"grad_norm": 0.6855290532112122,
|
| 1085 |
+
"learning_rate": 4.86e-05,
|
| 1086 |
+
"loss": 0.9658,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.056363636363636366,
|
| 1091 |
+
"grad_norm": 0.7574965357780457,
|
| 1092 |
+
"learning_rate": 4.859090909090909e-05,
|
| 1093 |
+
"loss": 0.9271,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.05672727272727273,
|
| 1098 |
+
"grad_norm": 0.6834359765052795,
|
| 1099 |
+
"learning_rate": 4.858181818181818e-05,
|
| 1100 |
+
"loss": 0.9395,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.05709090909090909,
|
| 1105 |
+
"grad_norm": 0.5964385271072388,
|
| 1106 |
+
"learning_rate": 4.857272727272727e-05,
|
| 1107 |
+
"loss": 1.0156,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.05745454545454545,
|
| 1112 |
+
"grad_norm": 0.6349005699157715,
|
| 1113 |
+
"learning_rate": 4.856363636363637e-05,
|
| 1114 |
+
"loss": 0.9195,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.05781818181818182,
|
| 1119 |
+
"grad_norm": 0.7871122360229492,
|
| 1120 |
+
"learning_rate": 4.855454545454546e-05,
|
| 1121 |
+
"loss": 1.008,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.05818181818181818,
|
| 1126 |
+
"grad_norm": 0.743977427482605,
|
| 1127 |
+
"learning_rate": 4.854545454545455e-05,
|
| 1128 |
+
"loss": 1.0277,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.058545454545454546,
|
| 1133 |
+
"grad_norm": 0.8460306525230408,
|
| 1134 |
+
"learning_rate": 4.853636363636364e-05,
|
| 1135 |
+
"loss": 1.0178,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.05890909090909091,
|
| 1140 |
+
"grad_norm": 0.6202099323272705,
|
| 1141 |
+
"learning_rate": 4.852727272727273e-05,
|
| 1142 |
+
"loss": 0.9564,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.059272727272727276,
|
| 1147 |
+
"grad_norm": 0.6819918155670166,
|
| 1148 |
+
"learning_rate": 4.851818181818182e-05,
|
| 1149 |
+
"loss": 1.008,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.05963636363636363,
|
| 1154 |
+
"grad_norm": 0.8113404512405396,
|
| 1155 |
+
"learning_rate": 4.850909090909091e-05,
|
| 1156 |
+
"loss": 1.0469,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.06,
|
| 1161 |
+
"grad_norm": 0.7554293274879456,
|
| 1162 |
+
"learning_rate": 4.85e-05,
|
| 1163 |
+
"loss": 0.9373,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.06036363636363636,
|
| 1168 |
+
"grad_norm": 0.7795326709747314,
|
| 1169 |
+
"learning_rate": 4.849090909090909e-05,
|
| 1170 |
+
"loss": 1.0633,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.06072727272727273,
|
| 1175 |
+
"grad_norm": 0.6483361721038818,
|
| 1176 |
+
"learning_rate": 4.848181818181819e-05,
|
| 1177 |
+
"loss": 1.0172,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.06109090909090909,
|
| 1182 |
+
"grad_norm": 0.7647920846939087,
|
| 1183 |
+
"learning_rate": 4.847272727272728e-05,
|
| 1184 |
+
"loss": 0.9861,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.061454545454545456,
|
| 1189 |
+
"grad_norm": 0.9443255066871643,
|
| 1190 |
+
"learning_rate": 4.846363636363637e-05,
|
| 1191 |
+
"loss": 1.0828,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.06181818181818182,
|
| 1196 |
+
"grad_norm": 0.8193866610527039,
|
| 1197 |
+
"learning_rate": 4.845454545454546e-05,
|
| 1198 |
+
"loss": 1.0604,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.06218181818181818,
|
| 1203 |
+
"grad_norm": 0.707897961139679,
|
| 1204 |
+
"learning_rate": 4.844545454545455e-05,
|
| 1205 |
+
"loss": 0.9975,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.06254545454545454,
|
| 1210 |
+
"grad_norm": 0.9864387512207031,
|
| 1211 |
+
"learning_rate": 4.8436363636363644e-05,
|
| 1212 |
+
"loss": 1.0488,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.06290909090909091,
|
| 1217 |
+
"grad_norm": 0.741073727607727,
|
| 1218 |
+
"learning_rate": 4.8427272727272734e-05,
|
| 1219 |
+
"loss": 0.9912,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.06327272727272727,
|
| 1224 |
+
"grad_norm": 0.6622138619422913,
|
| 1225 |
+
"learning_rate": 4.841818181818182e-05,
|
| 1226 |
+
"loss": 0.9904,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.06363636363636363,
|
| 1231 |
+
"grad_norm": 0.9334690570831299,
|
| 1232 |
+
"learning_rate": 4.840909090909091e-05,
|
| 1233 |
+
"loss": 0.9775,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.064,
|
| 1238 |
+
"grad_norm": 0.5777163505554199,
|
| 1239 |
+
"learning_rate": 4.8400000000000004e-05,
|
| 1240 |
+
"loss": 0.951,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.06436363636363636,
|
| 1245 |
+
"grad_norm": 0.9769054651260376,
|
| 1246 |
+
"learning_rate": 4.8390909090909094e-05,
|
| 1247 |
+
"loss": 0.9883,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.06472727272727273,
|
| 1252 |
+
"grad_norm": 0.8204796314239502,
|
| 1253 |
+
"learning_rate": 4.8381818181818184e-05,
|
| 1254 |
+
"loss": 0.9746,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.06509090909090909,
|
| 1259 |
+
"grad_norm": 0.8641470670700073,
|
| 1260 |
+
"learning_rate": 4.8372727272727274e-05,
|
| 1261 |
+
"loss": 1.0391,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.06545454545454546,
|
| 1266 |
+
"grad_norm": 0.732994794845581,
|
| 1267 |
+
"learning_rate": 4.8363636363636364e-05,
|
| 1268 |
+
"loss": 0.9494,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.06581818181818182,
|
| 1273 |
+
"grad_norm": 0.5632991194725037,
|
| 1274 |
+
"learning_rate": 4.835454545454546e-05,
|
| 1275 |
+
"loss": 1.1303,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.06618181818181817,
|
| 1280 |
+
"grad_norm": 0.6839861869812012,
|
| 1281 |
+
"learning_rate": 4.834545454545455e-05,
|
| 1282 |
+
"loss": 0.9652,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.06654545454545455,
|
| 1287 |
+
"grad_norm": 0.5407667756080627,
|
| 1288 |
+
"learning_rate": 4.833636363636364e-05,
|
| 1289 |
+
"loss": 0.9885,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.0669090909090909,
|
| 1294 |
+
"grad_norm": 0.8409451246261597,
|
| 1295 |
+
"learning_rate": 4.832727272727273e-05,
|
| 1296 |
+
"loss": 1.0197,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.06727272727272728,
|
| 1301 |
+
"grad_norm": 0.8460031151771545,
|
| 1302 |
+
"learning_rate": 4.8318181818181815e-05,
|
| 1303 |
+
"loss": 1.0137,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.06763636363636363,
|
| 1308 |
+
"grad_norm": 0.6154618859291077,
|
| 1309 |
+
"learning_rate": 4.830909090909091e-05,
|
| 1310 |
+
"loss": 0.9316,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.068,
|
| 1315 |
+
"grad_norm": 0.7288620471954346,
|
| 1316 |
+
"learning_rate": 4.83e-05,
|
| 1317 |
+
"loss": 0.957,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.06836363636363636,
|
| 1322 |
+
"grad_norm": 0.6734368205070496,
|
| 1323 |
+
"learning_rate": 4.829090909090909e-05,
|
| 1324 |
+
"loss": 0.9178,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.06872727272727273,
|
| 1329 |
+
"grad_norm": 0.5886709690093994,
|
| 1330 |
+
"learning_rate": 4.828181818181818e-05,
|
| 1331 |
+
"loss": 0.8715,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.06909090909090909,
|
| 1336 |
+
"grad_norm": 0.6626171469688416,
|
| 1337 |
+
"learning_rate": 4.827272727272727e-05,
|
| 1338 |
+
"loss": 1.0303,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.06945454545454545,
|
| 1343 |
+
"grad_norm": 0.6848061680793762,
|
| 1344 |
+
"learning_rate": 4.826363636363637e-05,
|
| 1345 |
+
"loss": 0.9477,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.06981818181818182,
|
| 1350 |
+
"grad_norm": 0.9979369640350342,
|
| 1351 |
+
"learning_rate": 4.825454545454546e-05,
|
| 1352 |
+
"loss": 1.1021,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.07018181818181818,
|
| 1357 |
+
"grad_norm": 0.8587698340415955,
|
| 1358 |
+
"learning_rate": 4.824545454545455e-05,
|
| 1359 |
+
"loss": 0.9975,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.07054545454545455,
|
| 1364 |
+
"grad_norm": 0.7505869269371033,
|
| 1365 |
+
"learning_rate": 4.823636363636364e-05,
|
| 1366 |
+
"loss": 0.902,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.07090909090909091,
|
| 1371 |
+
"grad_norm": 1.0160633325576782,
|
| 1372 |
+
"learning_rate": 4.822727272727273e-05,
|
| 1373 |
+
"loss": 1.0432,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.07127272727272728,
|
| 1378 |
+
"grad_norm": 0.6609899401664734,
|
| 1379 |
+
"learning_rate": 4.821818181818182e-05,
|
| 1380 |
+
"loss": 0.9576,
|
| 1381 |
+
"step": 1960
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.07163636363636364,
|
| 1385 |
+
"grad_norm": 0.8230269551277161,
|
| 1386 |
+
"learning_rate": 4.820909090909091e-05,
|
| 1387 |
+
"loss": 1.0172,
|
| 1388 |
+
"step": 1970
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.072,
|
| 1392 |
+
"grad_norm": 0.7711009979248047,
|
| 1393 |
+
"learning_rate": 4.82e-05,
|
| 1394 |
+
"loss": 1.133,
|
| 1395 |
+
"step": 1980
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.07236363636363637,
|
| 1399 |
+
"grad_norm": 0.8839887380599976,
|
| 1400 |
+
"learning_rate": 4.819090909090909e-05,
|
| 1401 |
+
"loss": 1.0662,
|
| 1402 |
+
"step": 1990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.07272727272727272,
|
| 1406 |
+
"grad_norm": 0.6672515869140625,
|
| 1407 |
+
"learning_rate": 4.8181818181818186e-05,
|
| 1408 |
+
"loss": 0.9764,
|
| 1409 |
+
"step": 2000
|
| 1410 |
+
}
|
| 1411 |
+
],
|
| 1412 |
+
"logging_steps": 10,
|
| 1413 |
+
"max_steps": 55000,
|
| 1414 |
+
"num_input_tokens_seen": 0,
|
| 1415 |
+
"num_train_epochs": 2,
|
| 1416 |
+
"save_steps": 1000,
|
| 1417 |
+
"stateful_callbacks": {
|
| 1418 |
+
"TrainerControl": {
|
| 1419 |
+
"args": {
|
| 1420 |
+
"should_epoch_stop": false,
|
| 1421 |
+
"should_evaluate": false,
|
| 1422 |
+
"should_log": false,
|
| 1423 |
+
"should_save": true,
|
| 1424 |
+
"should_training_stop": false
|
| 1425 |
+
},
|
| 1426 |
+
"attributes": {}
|
| 1427 |
+
}
|
| 1428 |
+
},
|
| 1429 |
+
"total_flos": 1.2566081788222917e+18,
|
| 1430 |
+
"train_batch_size": 2,
|
| 1431 |
+
"trial_name": null,
|
| 1432 |
+
"trial_params": null
|
| 1433 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ffab543a5a6285a5dc705d68751997721338129830541b8c103977c0c6dcf048
|
| 3 |
+
size 5368
|