Upload folder using huggingface_hub
Browse files- adapter_config.json +32 -0
- adapter_model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1013 -0
- training_args.bin +3 -0
adapter_config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "C:/Users/Eric/Desktop/repos/temp/GenAI-Mental-Health-train/training/Llama-2-7b-hf",
|
| 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": 8,
|
| 23 |
+
"rank_pattern": {},
|
| 24 |
+
"revision": null,
|
| 25 |
+
"target_modules": [
|
| 26 |
+
"v_proj",
|
| 27 |
+
"q_proj"
|
| 28 |
+
],
|
| 29 |
+
"task_type": "CAUSAL_LM",
|
| 30 |
+
"use_dora": false,
|
| 31 |
+
"use_rslora": false
|
| 32 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9910afeeb9c9206f1b2977f5666eb5e575166c48b1bb4b43d69aca4b6c46a887
|
| 3 |
+
size 16794200
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f3c3e76963019b08e902851d1ecd4e782b5edffe4aff2c2f6796b5a1d4de4f5
|
| 3 |
+
size 33662074
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d21e80acc1db437e2e013bb16826b3ed63dcdbe7f192acd3953c70574d06ab1
|
| 3 |
+
size 14244
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7f9e7345f008084f4ca2d26d07320162ec9d99156e386a48f4dc5753348e7a7d
|
| 3 |
+
size 988
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f7ad8c2dbb2e4c2b467f1f286e0881a55b5a4ad900e2ef4212fe6617e5f5be73
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1013 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.609450841259822,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 14000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.004353220294713014,
|
| 13 |
+
"grad_norm": 0.44108787178993225,
|
| 14 |
+
"learning_rate": 4.000000000000001e-06,
|
| 15 |
+
"loss": 1.3221,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.008706440589426028,
|
| 20 |
+
"grad_norm": 0.538872241973877,
|
| 21 |
+
"learning_rate": 8.000000000000001e-06,
|
| 22 |
+
"loss": 1.3127,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.013059660884139042,
|
| 27 |
+
"grad_norm": 0.7150405645370483,
|
| 28 |
+
"learning_rate": 1.2e-05,
|
| 29 |
+
"loss": 1.283,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.017412881178852056,
|
| 34 |
+
"grad_norm": 0.6822145581245422,
|
| 35 |
+
"learning_rate": 1.5960000000000003e-05,
|
| 36 |
+
"loss": 1.2565,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.02176610147356507,
|
| 41 |
+
"grad_norm": 0.8248649835586548,
|
| 42 |
+
"learning_rate": 1.9960000000000002e-05,
|
| 43 |
+
"loss": 1.2557,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.026119321768278084,
|
| 48 |
+
"grad_norm": 0.7033706307411194,
|
| 49 |
+
"learning_rate": 1.9982685496917496e-05,
|
| 50 |
+
"loss": 1.2316,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.030472542062991097,
|
| 55 |
+
"grad_norm": 0.8667769432067871,
|
| 56 |
+
"learning_rate": 1.996519609986446e-05,
|
| 57 |
+
"loss": 1.2398,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03482576235770411,
|
| 62 |
+
"grad_norm": 0.8944305777549744,
|
| 63 |
+
"learning_rate": 1.9947706702811424e-05,
|
| 64 |
+
"loss": 1.2384,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03917898265241713,
|
| 69 |
+
"grad_norm": 0.8540185689926147,
|
| 70 |
+
"learning_rate": 1.9930217305758386e-05,
|
| 71 |
+
"loss": 1.244,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.04353220294713014,
|
| 76 |
+
"grad_norm": 0.8816583752632141,
|
| 77 |
+
"learning_rate": 1.991272790870535e-05,
|
| 78 |
+
"loss": 1.2539,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04788542324184315,
|
| 83 |
+
"grad_norm": 0.8317933082580566,
|
| 84 |
+
"learning_rate": 1.989523851165231e-05,
|
| 85 |
+
"loss": 1.2407,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.05223864353655617,
|
| 90 |
+
"grad_norm": 0.9426454305648804,
|
| 91 |
+
"learning_rate": 1.9877749114599277e-05,
|
| 92 |
+
"loss": 1.2257,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.056591863831269185,
|
| 97 |
+
"grad_norm": 1.0447773933410645,
|
| 98 |
+
"learning_rate": 1.986025971754624e-05,
|
| 99 |
+
"loss": 1.2454,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.060945084125982193,
|
| 104 |
+
"grad_norm": 0.8134283423423767,
|
| 105 |
+
"learning_rate": 1.9842770320493202e-05,
|
| 106 |
+
"loss": 1.2591,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.0652983044206952,
|
| 111 |
+
"grad_norm": 0.9173294901847839,
|
| 112 |
+
"learning_rate": 1.9825280923440168e-05,
|
| 113 |
+
"loss": 1.2322,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.06965152471540823,
|
| 118 |
+
"grad_norm": 0.8239843845367432,
|
| 119 |
+
"learning_rate": 1.980779152638713e-05,
|
| 120 |
+
"loss": 1.2309,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.07400474501012123,
|
| 125 |
+
"grad_norm": 0.935463011264801,
|
| 126 |
+
"learning_rate": 1.9790302129334092e-05,
|
| 127 |
+
"loss": 1.223,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.07835796530483426,
|
| 132 |
+
"grad_norm": 0.9278773665428162,
|
| 133 |
+
"learning_rate": 1.9772812732281058e-05,
|
| 134 |
+
"loss": 1.2402,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.08271118559954727,
|
| 139 |
+
"grad_norm": 0.7804251313209534,
|
| 140 |
+
"learning_rate": 1.975532333522802e-05,
|
| 141 |
+
"loss": 1.2243,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.08706440589426027,
|
| 146 |
+
"grad_norm": 0.9013062715530396,
|
| 147 |
+
"learning_rate": 1.9737833938174983e-05,
|
| 148 |
+
"loss": 1.2304,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.0914176261889733,
|
| 153 |
+
"grad_norm": 1.1383230686187744,
|
| 154 |
+
"learning_rate": 1.972034454112195e-05,
|
| 155 |
+
"loss": 1.2341,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.0957708464836863,
|
| 160 |
+
"grad_norm": 0.8813136219978333,
|
| 161 |
+
"learning_rate": 1.970285514406891e-05,
|
| 162 |
+
"loss": 1.2021,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.10012406677839931,
|
| 167 |
+
"grad_norm": 0.9946884512901306,
|
| 168 |
+
"learning_rate": 1.9685365747015873e-05,
|
| 169 |
+
"loss": 1.2164,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.10447728707311234,
|
| 174 |
+
"grad_norm": 0.9189836978912354,
|
| 175 |
+
"learning_rate": 1.9667876349962836e-05,
|
| 176 |
+
"loss": 1.2249,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.10883050736782535,
|
| 181 |
+
"grad_norm": 0.9713960886001587,
|
| 182 |
+
"learning_rate": 1.9650386952909798e-05,
|
| 183 |
+
"loss": 1.2323,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.11318372766253837,
|
| 188 |
+
"grad_norm": 0.866607666015625,
|
| 189 |
+
"learning_rate": 1.9632897555856764e-05,
|
| 190 |
+
"loss": 1.2091,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.11753694795725138,
|
| 195 |
+
"grad_norm": 0.8033185005187988,
|
| 196 |
+
"learning_rate": 1.9615408158803726e-05,
|
| 197 |
+
"loss": 1.2112,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.12189016825196439,
|
| 202 |
+
"grad_norm": 1.0787014961242676,
|
| 203 |
+
"learning_rate": 1.959791876175069e-05,
|
| 204 |
+
"loss": 1.2105,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.1262433885466774,
|
| 209 |
+
"grad_norm": 0.9159247875213623,
|
| 210 |
+
"learning_rate": 1.9580429364697654e-05,
|
| 211 |
+
"loss": 1.2168,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.1305966088413904,
|
| 216 |
+
"grad_norm": 0.9194386601448059,
|
| 217 |
+
"learning_rate": 1.9562939967644617e-05,
|
| 218 |
+
"loss": 1.2225,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.13494982913610343,
|
| 223 |
+
"grad_norm": 1.0548698902130127,
|
| 224 |
+
"learning_rate": 1.954545057059158e-05,
|
| 225 |
+
"loss": 1.2157,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.13930304943081645,
|
| 230 |
+
"grad_norm": 1.23219633102417,
|
| 231 |
+
"learning_rate": 1.952796117353854e-05,
|
| 232 |
+
"loss": 1.2325,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.14365626972552947,
|
| 237 |
+
"grad_norm": 1.0198304653167725,
|
| 238 |
+
"learning_rate": 1.9510471776485507e-05,
|
| 239 |
+
"loss": 1.1953,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.14800949002024247,
|
| 244 |
+
"grad_norm": 1.0076764822006226,
|
| 245 |
+
"learning_rate": 1.949298237943247e-05,
|
| 246 |
+
"loss": 1.2096,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.1523627103149555,
|
| 251 |
+
"grad_norm": 0.8188544511795044,
|
| 252 |
+
"learning_rate": 1.9475492982379432e-05,
|
| 253 |
+
"loss": 1.2237,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1567159306096685,
|
| 258 |
+
"grad_norm": 0.8744425177574158,
|
| 259 |
+
"learning_rate": 1.9458003585326398e-05,
|
| 260 |
+
"loss": 1.2166,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.1610691509043815,
|
| 265 |
+
"grad_norm": 0.9896447062492371,
|
| 266 |
+
"learning_rate": 1.944051418827336e-05,
|
| 267 |
+
"loss": 1.2006,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.16542237119909453,
|
| 272 |
+
"grad_norm": 1.201147198677063,
|
| 273 |
+
"learning_rate": 1.9423024791220323e-05,
|
| 274 |
+
"loss": 1.2062,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.16977559149380755,
|
| 279 |
+
"grad_norm": 1.0766533613204956,
|
| 280 |
+
"learning_rate": 1.940553539416729e-05,
|
| 281 |
+
"loss": 1.2218,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.17412881178852055,
|
| 286 |
+
"grad_norm": 1.007964015007019,
|
| 287 |
+
"learning_rate": 1.938804599711425e-05,
|
| 288 |
+
"loss": 1.2343,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.17848203208323357,
|
| 293 |
+
"grad_norm": 1.0383857488632202,
|
| 294 |
+
"learning_rate": 1.9370556600061213e-05,
|
| 295 |
+
"loss": 1.235,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1828352523779466,
|
| 300 |
+
"grad_norm": 1.0007116794586182,
|
| 301 |
+
"learning_rate": 1.935306720300818e-05,
|
| 302 |
+
"loss": 1.2158,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1871884726726596,
|
| 307 |
+
"grad_norm": 0.8174022436141968,
|
| 308 |
+
"learning_rate": 1.933557780595514e-05,
|
| 309 |
+
"loss": 1.2093,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.1915416929673726,
|
| 314 |
+
"grad_norm": 1.0221714973449707,
|
| 315 |
+
"learning_rate": 1.9318088408902104e-05,
|
| 316 |
+
"loss": 1.2085,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.19589491326208563,
|
| 321 |
+
"grad_norm": 0.8928303718566895,
|
| 322 |
+
"learning_rate": 1.930059901184907e-05,
|
| 323 |
+
"loss": 1.2233,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.20024813355679863,
|
| 328 |
+
"grad_norm": 0.8953472375869751,
|
| 329 |
+
"learning_rate": 1.9283109614796032e-05,
|
| 330 |
+
"loss": 1.2148,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.20460135385151165,
|
| 335 |
+
"grad_norm": 1.1339576244354248,
|
| 336 |
+
"learning_rate": 1.9265620217742994e-05,
|
| 337 |
+
"loss": 1.2164,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.20895457414622468,
|
| 342 |
+
"grad_norm": 1.027590036392212,
|
| 343 |
+
"learning_rate": 1.924813082068996e-05,
|
| 344 |
+
"loss": 1.1952,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.21330779444093767,
|
| 349 |
+
"grad_norm": 0.8583177328109741,
|
| 350 |
+
"learning_rate": 1.9230641423636922e-05,
|
| 351 |
+
"loss": 1.2158,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.2176610147356507,
|
| 356 |
+
"grad_norm": 1.2040601968765259,
|
| 357 |
+
"learning_rate": 1.9213152026583885e-05,
|
| 358 |
+
"loss": 1.2083,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.22201423503036372,
|
| 363 |
+
"grad_norm": 1.1134988069534302,
|
| 364 |
+
"learning_rate": 1.919566262953085e-05,
|
| 365 |
+
"loss": 1.2196,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.22636745532507674,
|
| 370 |
+
"grad_norm": 0.8359972238540649,
|
| 371 |
+
"learning_rate": 1.9178173232477813e-05,
|
| 372 |
+
"loss": 1.2122,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.23072067561978973,
|
| 377 |
+
"grad_norm": 0.930724024772644,
|
| 378 |
+
"learning_rate": 1.9160683835424775e-05,
|
| 379 |
+
"loss": 1.2108,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.23507389591450276,
|
| 384 |
+
"grad_norm": 0.9618527889251709,
|
| 385 |
+
"learning_rate": 1.914319443837174e-05,
|
| 386 |
+
"loss": 1.201,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.23942711620921578,
|
| 391 |
+
"grad_norm": 0.9389538764953613,
|
| 392 |
+
"learning_rate": 1.9125705041318703e-05,
|
| 393 |
+
"loss": 1.1959,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.24378033650392877,
|
| 398 |
+
"grad_norm": 1.0064427852630615,
|
| 399 |
+
"learning_rate": 1.9108215644265666e-05,
|
| 400 |
+
"loss": 1.2187,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.2481335567986418,
|
| 405 |
+
"grad_norm": 0.9333221912384033,
|
| 406 |
+
"learning_rate": 1.9090726247212628e-05,
|
| 407 |
+
"loss": 1.2114,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.2524867770933548,
|
| 412 |
+
"grad_norm": 0.9797629714012146,
|
| 413 |
+
"learning_rate": 1.9073236850159594e-05,
|
| 414 |
+
"loss": 1.2075,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.25683999738806784,
|
| 419 |
+
"grad_norm": 0.9818430542945862,
|
| 420 |
+
"learning_rate": 1.9055747453106556e-05,
|
| 421 |
+
"loss": 1.2108,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.2611932176827808,
|
| 426 |
+
"grad_norm": 0.9731236696243286,
|
| 427 |
+
"learning_rate": 1.903825805605352e-05,
|
| 428 |
+
"loss": 1.2101,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.26554643797749383,
|
| 433 |
+
"grad_norm": 0.9190470576286316,
|
| 434 |
+
"learning_rate": 1.902076865900048e-05,
|
| 435 |
+
"loss": 1.2158,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.26989965827220685,
|
| 440 |
+
"grad_norm": 0.9030302166938782,
|
| 441 |
+
"learning_rate": 1.9003279261947443e-05,
|
| 442 |
+
"loss": 1.2098,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.2742528785669199,
|
| 447 |
+
"grad_norm": 1.0309573411941528,
|
| 448 |
+
"learning_rate": 1.898578986489441e-05,
|
| 449 |
+
"loss": 1.2134,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.2786060988616329,
|
| 454 |
+
"grad_norm": 0.8960947394371033,
|
| 455 |
+
"learning_rate": 1.896830046784137e-05,
|
| 456 |
+
"loss": 1.2051,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.2829593191563459,
|
| 461 |
+
"grad_norm": 1.074310302734375,
|
| 462 |
+
"learning_rate": 1.8950811070788334e-05,
|
| 463 |
+
"loss": 1.2012,
|
| 464 |
+
"step": 6500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.28731253945105895,
|
| 468 |
+
"grad_norm": 0.9871541261672974,
|
| 469 |
+
"learning_rate": 1.89333216737353e-05,
|
| 470 |
+
"loss": 1.2157,
|
| 471 |
+
"step": 6600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2916657597457719,
|
| 475 |
+
"grad_norm": 0.923338770866394,
|
| 476 |
+
"learning_rate": 1.8915832276682262e-05,
|
| 477 |
+
"loss": 1.1937,
|
| 478 |
+
"step": 6700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.29601898004048494,
|
| 482 |
+
"grad_norm": 1.086458683013916,
|
| 483 |
+
"learning_rate": 1.8898342879629225e-05,
|
| 484 |
+
"loss": 1.1976,
|
| 485 |
+
"step": 6800
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.30037220033519796,
|
| 489 |
+
"grad_norm": 0.9264793992042542,
|
| 490 |
+
"learning_rate": 1.888085348257619e-05,
|
| 491 |
+
"loss": 1.1976,
|
| 492 |
+
"step": 6900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.304725420629911,
|
| 496 |
+
"grad_norm": 0.9164808392524719,
|
| 497 |
+
"learning_rate": 1.8863364085523153e-05,
|
| 498 |
+
"loss": 1.2033,
|
| 499 |
+
"step": 7000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.309078640924624,
|
| 503 |
+
"grad_norm": 0.9287434816360474,
|
| 504 |
+
"learning_rate": 1.8845874688470115e-05,
|
| 505 |
+
"loss": 1.2307,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.313431861219337,
|
| 510 |
+
"grad_norm": 0.8909704685211182,
|
| 511 |
+
"learning_rate": 1.882838529141708e-05,
|
| 512 |
+
"loss": 1.2216,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.31778508151405,
|
| 517 |
+
"grad_norm": 1.1148613691329956,
|
| 518 |
+
"learning_rate": 1.8810895894364043e-05,
|
| 519 |
+
"loss": 1.209,
|
| 520 |
+
"step": 7300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.322138301808763,
|
| 524 |
+
"grad_norm": 0.9600721597671509,
|
| 525 |
+
"learning_rate": 1.8793406497311006e-05,
|
| 526 |
+
"loss": 1.1872,
|
| 527 |
+
"step": 7400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.32649152210347604,
|
| 531 |
+
"grad_norm": 0.9834184646606445,
|
| 532 |
+
"learning_rate": 1.877591710025797e-05,
|
| 533 |
+
"loss": 1.2136,
|
| 534 |
+
"step": 7500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.33084474239818906,
|
| 538 |
+
"grad_norm": 1.0687185525894165,
|
| 539 |
+
"learning_rate": 1.8758427703204934e-05,
|
| 540 |
+
"loss": 1.199,
|
| 541 |
+
"step": 7600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3351979626929021,
|
| 545 |
+
"grad_norm": 1.1417192220687866,
|
| 546 |
+
"learning_rate": 1.8740938306151896e-05,
|
| 547 |
+
"loss": 1.2027,
|
| 548 |
+
"step": 7700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.3395511829876151,
|
| 552 |
+
"grad_norm": 0.935863196849823,
|
| 553 |
+
"learning_rate": 1.8723448909098862e-05,
|
| 554 |
+
"loss": 1.2047,
|
| 555 |
+
"step": 7800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.3439044032823281,
|
| 559 |
+
"grad_norm": 0.9869050979614258,
|
| 560 |
+
"learning_rate": 1.8705959512045824e-05,
|
| 561 |
+
"loss": 1.2147,
|
| 562 |
+
"step": 7900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.3482576235770411,
|
| 566 |
+
"grad_norm": 0.9544995427131653,
|
| 567 |
+
"learning_rate": 1.8688470114992787e-05,
|
| 568 |
+
"loss": 1.1976,
|
| 569 |
+
"step": 8000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.3526108438717541,
|
| 573 |
+
"grad_norm": 0.9217939376831055,
|
| 574 |
+
"learning_rate": 1.867098071793975e-05,
|
| 575 |
+
"loss": 1.206,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.35696406416646714,
|
| 580 |
+
"grad_norm": 1.1815359592437744,
|
| 581 |
+
"learning_rate": 1.8653491320886715e-05,
|
| 582 |
+
"loss": 1.1999,
|
| 583 |
+
"step": 8200
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.36131728446118017,
|
| 587 |
+
"grad_norm": 1.1035760641098022,
|
| 588 |
+
"learning_rate": 1.863617681780421e-05,
|
| 589 |
+
"loss": 1.2166,
|
| 590 |
+
"step": 8300
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3656705047558932,
|
| 594 |
+
"grad_norm": 1.0326738357543945,
|
| 595 |
+
"learning_rate": 1.861868742075117e-05,
|
| 596 |
+
"loss": 1.2088,
|
| 597 |
+
"step": 8400
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.3700237250506062,
|
| 601 |
+
"grad_norm": 1.3306974172592163,
|
| 602 |
+
"learning_rate": 1.8601198023698134e-05,
|
| 603 |
+
"loss": 1.2078,
|
| 604 |
+
"step": 8500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.3743769453453192,
|
| 608 |
+
"grad_norm": 1.0832709074020386,
|
| 609 |
+
"learning_rate": 1.85837086266451e-05,
|
| 610 |
+
"loss": 1.2013,
|
| 611 |
+
"step": 8600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.3787301656400322,
|
| 615 |
+
"grad_norm": 0.9759966135025024,
|
| 616 |
+
"learning_rate": 1.8566219229592062e-05,
|
| 617 |
+
"loss": 1.1938,
|
| 618 |
+
"step": 8700
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.3830833859347452,
|
| 622 |
+
"grad_norm": 1.15034818649292,
|
| 623 |
+
"learning_rate": 1.8548729832539024e-05,
|
| 624 |
+
"loss": 1.2191,
|
| 625 |
+
"step": 8800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.38743660622945825,
|
| 629 |
+
"grad_norm": 0.9838187098503113,
|
| 630 |
+
"learning_rate": 1.853124043548599e-05,
|
| 631 |
+
"loss": 1.2134,
|
| 632 |
+
"step": 8900
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.39178982652417127,
|
| 636 |
+
"grad_norm": 1.2600586414337158,
|
| 637 |
+
"learning_rate": 1.8513751038432953e-05,
|
| 638 |
+
"loss": 1.2171,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3961430468188843,
|
| 643 |
+
"grad_norm": 0.9450699090957642,
|
| 644 |
+
"learning_rate": 1.8496261641379915e-05,
|
| 645 |
+
"loss": 1.1984,
|
| 646 |
+
"step": 9100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.40049626711359726,
|
| 650 |
+
"grad_norm": 0.9582616090774536,
|
| 651 |
+
"learning_rate": 1.847877224432688e-05,
|
| 652 |
+
"loss": 1.2142,
|
| 653 |
+
"step": 9200
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.4048494874083103,
|
| 657 |
+
"grad_norm": 0.9603105783462524,
|
| 658 |
+
"learning_rate": 1.8461282847273843e-05,
|
| 659 |
+
"loss": 1.2079,
|
| 660 |
+
"step": 9300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.4092027077030233,
|
| 664 |
+
"grad_norm": 1.0588648319244385,
|
| 665 |
+
"learning_rate": 1.8443793450220805e-05,
|
| 666 |
+
"loss": 1.2045,
|
| 667 |
+
"step": 9400
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.41355592799773633,
|
| 671 |
+
"grad_norm": 0.9764322638511658,
|
| 672 |
+
"learning_rate": 1.8426304053167768e-05,
|
| 673 |
+
"loss": 1.2056,
|
| 674 |
+
"step": 9500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.41790914829244935,
|
| 678 |
+
"grad_norm": 0.958659291267395,
|
| 679 |
+
"learning_rate": 1.8408814656114734e-05,
|
| 680 |
+
"loss": 1.2049,
|
| 681 |
+
"step": 9600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.4222623685871624,
|
| 685 |
+
"grad_norm": 0.9048427939414978,
|
| 686 |
+
"learning_rate": 1.8391325259061696e-05,
|
| 687 |
+
"loss": 1.2104,
|
| 688 |
+
"step": 9700
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.42661558888187534,
|
| 692 |
+
"grad_norm": 0.9958964586257935,
|
| 693 |
+
"learning_rate": 1.837383586200866e-05,
|
| 694 |
+
"loss": 1.2112,
|
| 695 |
+
"step": 9800
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.43096880917658836,
|
| 699 |
+
"grad_norm": 1.0927664041519165,
|
| 700 |
+
"learning_rate": 1.835634646495562e-05,
|
| 701 |
+
"loss": 1.1961,
|
| 702 |
+
"step": 9900
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.4353220294713014,
|
| 706 |
+
"grad_norm": 1.031762719154358,
|
| 707 |
+
"learning_rate": 1.8338857067902583e-05,
|
| 708 |
+
"loss": 1.2086,
|
| 709 |
+
"step": 10000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.4396752497660144,
|
| 713 |
+
"grad_norm": 1.1390494108200073,
|
| 714 |
+
"learning_rate": 1.832136767084955e-05,
|
| 715 |
+
"loss": 1.1972,
|
| 716 |
+
"step": 10100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.44402847006072743,
|
| 720 |
+
"grad_norm": 1.0771234035491943,
|
| 721 |
+
"learning_rate": 1.830387827379651e-05,
|
| 722 |
+
"loss": 1.2075,
|
| 723 |
+
"step": 10200
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.44838169035544045,
|
| 727 |
+
"grad_norm": 1.0366460084915161,
|
| 728 |
+
"learning_rate": 1.8286388876743474e-05,
|
| 729 |
+
"loss": 1.2099,
|
| 730 |
+
"step": 10300
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.4527349106501535,
|
| 734 |
+
"grad_norm": 1.2623414993286133,
|
| 735 |
+
"learning_rate": 1.826889947969044e-05,
|
| 736 |
+
"loss": 1.1907,
|
| 737 |
+
"step": 10400
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.45708813094486644,
|
| 741 |
+
"grad_norm": 0.9830704927444458,
|
| 742 |
+
"learning_rate": 1.8251410082637402e-05,
|
| 743 |
+
"loss": 1.2053,
|
| 744 |
+
"step": 10500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.46144135123957947,
|
| 748 |
+
"grad_norm": 0.9676218032836914,
|
| 749 |
+
"learning_rate": 1.8233920685584364e-05,
|
| 750 |
+
"loss": 1.1842,
|
| 751 |
+
"step": 10600
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.4657945715342925,
|
| 755 |
+
"grad_norm": 1.2105963230133057,
|
| 756 |
+
"learning_rate": 1.821643128853133e-05,
|
| 757 |
+
"loss": 1.1837,
|
| 758 |
+
"step": 10700
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.4701477918290055,
|
| 762 |
+
"grad_norm": 1.0238980054855347,
|
| 763 |
+
"learning_rate": 1.8198941891478292e-05,
|
| 764 |
+
"loss": 1.1819,
|
| 765 |
+
"step": 10800
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.47450101212371854,
|
| 769 |
+
"grad_norm": 0.9666149020195007,
|
| 770 |
+
"learning_rate": 1.8181452494425255e-05,
|
| 771 |
+
"loss": 1.1948,
|
| 772 |
+
"step": 10900
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.47885423241843156,
|
| 776 |
+
"grad_norm": 1.0068769454956055,
|
| 777 |
+
"learning_rate": 1.816396309737222e-05,
|
| 778 |
+
"loss": 1.1871,
|
| 779 |
+
"step": 11000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.4832074527131445,
|
| 783 |
+
"grad_norm": 0.8864049315452576,
|
| 784 |
+
"learning_rate": 1.8146473700319183e-05,
|
| 785 |
+
"loss": 1.224,
|
| 786 |
+
"step": 11100
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.48756067300785755,
|
| 790 |
+
"grad_norm": 0.8751236796379089,
|
| 791 |
+
"learning_rate": 1.8128984303266145e-05,
|
| 792 |
+
"loss": 1.2245,
|
| 793 |
+
"step": 11200
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.49191389330257057,
|
| 797 |
+
"grad_norm": 1.1171832084655762,
|
| 798 |
+
"learning_rate": 1.811149490621311e-05,
|
| 799 |
+
"loss": 1.1964,
|
| 800 |
+
"step": 11300
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4962671135972836,
|
| 804 |
+
"grad_norm": 0.9493364095687866,
|
| 805 |
+
"learning_rate": 1.8094005509160073e-05,
|
| 806 |
+
"loss": 1.2098,
|
| 807 |
+
"step": 11400
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.5006203338919966,
|
| 811 |
+
"grad_norm": 1.0340662002563477,
|
| 812 |
+
"learning_rate": 1.8076516112107036e-05,
|
| 813 |
+
"loss": 1.1899,
|
| 814 |
+
"step": 11500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.5049735541867096,
|
| 818 |
+
"grad_norm": 0.9075784683227539,
|
| 819 |
+
"learning_rate": 1.8059026715054e-05,
|
| 820 |
+
"loss": 1.2088,
|
| 821 |
+
"step": 11600
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.5093267744814226,
|
| 825 |
+
"grad_norm": 0.9786097407341003,
|
| 826 |
+
"learning_rate": 1.8041537318000964e-05,
|
| 827 |
+
"loss": 1.2074,
|
| 828 |
+
"step": 11700
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.5136799947761357,
|
| 832 |
+
"grad_norm": 0.8463122248649597,
|
| 833 |
+
"learning_rate": 1.8024047920947926e-05,
|
| 834 |
+
"loss": 1.2007,
|
| 835 |
+
"step": 11800
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.5180332150708487,
|
| 839 |
+
"grad_norm": 1.4936442375183105,
|
| 840 |
+
"learning_rate": 1.800655852389489e-05,
|
| 841 |
+
"loss": 1.196,
|
| 842 |
+
"step": 11900
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.5223864353655616,
|
| 846 |
+
"grad_norm": 0.9578948616981506,
|
| 847 |
+
"learning_rate": 1.7989069126841854e-05,
|
| 848 |
+
"loss": 1.2106,
|
| 849 |
+
"step": 12000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.5267396556602747,
|
| 853 |
+
"grad_norm": 1.142462968826294,
|
| 854 |
+
"learning_rate": 1.7971579729788817e-05,
|
| 855 |
+
"loss": 1.1811,
|
| 856 |
+
"step": 12100
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.5310928759549877,
|
| 860 |
+
"grad_norm": 0.9496989846229553,
|
| 861 |
+
"learning_rate": 1.795409033273578e-05,
|
| 862 |
+
"loss": 1.2012,
|
| 863 |
+
"step": 12200
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.5354460962497007,
|
| 867 |
+
"grad_norm": 1.2966980934143066,
|
| 868 |
+
"learning_rate": 1.7936775829653274e-05,
|
| 869 |
+
"loss": 1.1976,
|
| 870 |
+
"step": 12300
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.5397993165444137,
|
| 874 |
+
"grad_norm": 0.9380930066108704,
|
| 875 |
+
"learning_rate": 1.791928643260024e-05,
|
| 876 |
+
"loss": 1.1912,
|
| 877 |
+
"step": 12400
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.5441525368391268,
|
| 881 |
+
"grad_norm": 1.1202783584594727,
|
| 882 |
+
"learning_rate": 1.7901797035547202e-05,
|
| 883 |
+
"loss": 1.2141,
|
| 884 |
+
"step": 12500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.5485057571338398,
|
| 888 |
+
"grad_norm": 0.8347093462944031,
|
| 889 |
+
"learning_rate": 1.7884307638494164e-05,
|
| 890 |
+
"loss": 1.1889,
|
| 891 |
+
"step": 12600
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.5528589774285527,
|
| 895 |
+
"grad_norm": 1.0347511768341064,
|
| 896 |
+
"learning_rate": 1.786681824144113e-05,
|
| 897 |
+
"loss": 1.1886,
|
| 898 |
+
"step": 12700
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.5572121977232658,
|
| 902 |
+
"grad_norm": 1.3284380435943604,
|
| 903 |
+
"learning_rate": 1.7849328844388092e-05,
|
| 904 |
+
"loss": 1.2096,
|
| 905 |
+
"step": 12800
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.5615654180179788,
|
| 909 |
+
"grad_norm": 1.0298584699630737,
|
| 910 |
+
"learning_rate": 1.7831839447335055e-05,
|
| 911 |
+
"loss": 1.1839,
|
| 912 |
+
"step": 12900
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.5659186383126918,
|
| 916 |
+
"grad_norm": 0.9223812222480774,
|
| 917 |
+
"learning_rate": 1.781435005028202e-05,
|
| 918 |
+
"loss": 1.1512,
|
| 919 |
+
"step": 13000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.5702718586074048,
|
| 923 |
+
"grad_norm": 0.9801159501075745,
|
| 924 |
+
"learning_rate": 1.7796860653228983e-05,
|
| 925 |
+
"loss": 1.2046,
|
| 926 |
+
"step": 13100
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.5746250789021179,
|
| 930 |
+
"grad_norm": 1.2689110040664673,
|
| 931 |
+
"learning_rate": 1.7779371256175945e-05,
|
| 932 |
+
"loss": 1.1929,
|
| 933 |
+
"step": 13200
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.5789782991968309,
|
| 937 |
+
"grad_norm": 1.035518765449524,
|
| 938 |
+
"learning_rate": 1.7761881859122908e-05,
|
| 939 |
+
"loss": 1.1781,
|
| 940 |
+
"step": 13300
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.5833315194915438,
|
| 944 |
+
"grad_norm": 0.8404269814491272,
|
| 945 |
+
"learning_rate": 1.7744392462069873e-05,
|
| 946 |
+
"loss": 1.2005,
|
| 947 |
+
"step": 13400
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5876847397862569,
|
| 951 |
+
"grad_norm": 1.1780256032943726,
|
| 952 |
+
"learning_rate": 1.7726903065016832e-05,
|
| 953 |
+
"loss": 1.2082,
|
| 954 |
+
"step": 13500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.5920379600809699,
|
| 958 |
+
"grad_norm": 1.147164225578308,
|
| 959 |
+
"learning_rate": 1.7709413667963798e-05,
|
| 960 |
+
"loss": 1.2121,
|
| 961 |
+
"step": 13600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.596391180375683,
|
| 965 |
+
"grad_norm": 1.1297084093093872,
|
| 966 |
+
"learning_rate": 1.769192427091076e-05,
|
| 967 |
+
"loss": 1.1875,
|
| 968 |
+
"step": 13700
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.6007444006703959,
|
| 972 |
+
"grad_norm": 0.9906993508338928,
|
| 973 |
+
"learning_rate": 1.7674434873857723e-05,
|
| 974 |
+
"loss": 1.196,
|
| 975 |
+
"step": 13800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.6050976209651089,
|
| 979 |
+
"grad_norm": 0.8798189163208008,
|
| 980 |
+
"learning_rate": 1.765694547680469e-05,
|
| 981 |
+
"loss": 1.1953,
|
| 982 |
+
"step": 13900
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.609450841259822,
|
| 986 |
+
"grad_norm": 0.9738469123840332,
|
| 987 |
+
"learning_rate": 1.763945607975165e-05,
|
| 988 |
+
"loss": 1.1863,
|
| 989 |
+
"step": 14000
|
| 990 |
+
}
|
| 991 |
+
],
|
| 992 |
+
"logging_steps": 100,
|
| 993 |
+
"max_steps": 114855,
|
| 994 |
+
"num_input_tokens_seen": 0,
|
| 995 |
+
"num_train_epochs": 5,
|
| 996 |
+
"save_steps": 1000,
|
| 997 |
+
"stateful_callbacks": {
|
| 998 |
+
"TrainerControl": {
|
| 999 |
+
"args": {
|
| 1000 |
+
"should_epoch_stop": false,
|
| 1001 |
+
"should_evaluate": false,
|
| 1002 |
+
"should_log": false,
|
| 1003 |
+
"should_save": true,
|
| 1004 |
+
"should_training_stop": false
|
| 1005 |
+
},
|
| 1006 |
+
"attributes": {}
|
| 1007 |
+
}
|
| 1008 |
+
},
|
| 1009 |
+
"total_flos": 2.1684380318642995e+18,
|
| 1010 |
+
"train_batch_size": 2,
|
| 1011 |
+
"trial_name": null,
|
| 1012 |
+
"trial_params": null
|
| 1013 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:08fd64a4d6d33316874022441268fbff01bc2dd77c08cfc1d7d7786a74f405aa
|
| 3 |
+
size 5304
|