Initial commit with LoRA
Browse files- adapter_config.json +26 -0
- adapter_model.safetensors +3 -0
- generation_config.json +12 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1230 -0
- training_args.bin +3 -0
adapter_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "HuggingFaceM4/Idefics3-8B-Llama3",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": false,
|
| 8 |
+
"init_lora_weights": "gaussian",
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 8,
|
| 14 |
+
"lora_dropout": 0.1,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 8,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": ".*(text_model|connector).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$",
|
| 23 |
+
"task_type": null,
|
| 24 |
+
"use_dora": false,
|
| 25 |
+
"use_rslora": false
|
| 26 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c3cd0b773b95893136e29473a57dad00ab4d9cea5ce1636d500aabf22def7bff
|
| 3 |
+
size 83950224
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 128000,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
128001,
|
| 6 |
+
128008,
|
| 7 |
+
128009
|
| 8 |
+
],
|
| 9 |
+
"pad_token_id": 128002,
|
| 10 |
+
"transformers_version": "4.45.0.dev0",
|
| 11 |
+
"use_cache": false
|
| 12 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5661cc038018142132cb5f7dcf941a2ff396fdff16decdf361a0c946ca872f44
|
| 3 |
+
size 42546196
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fbcbb0c40771333f08d627ee7a2b3454868c252ac0f43546dc445586a56c2a87
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:097ce7a83bd4f3af5436d8b17f963920c1d8c472a5e334590645c58bb31b38dd
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1230 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 4287,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.005831583858175881,
|
| 13 |
+
"grad_norm": 1.3819491863250732,
|
| 14 |
+
"learning_rate": 5e-05,
|
| 15 |
+
"loss": 8.7045,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.011663167716351761,
|
| 20 |
+
"grad_norm": 0.21646396815776825,
|
| 21 |
+
"learning_rate": 0.0001,
|
| 22 |
+
"loss": 0.0321,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01749475157452764,
|
| 27 |
+
"grad_norm": 0.11650694161653519,
|
| 28 |
+
"learning_rate": 9.940995987727166e-05,
|
| 29 |
+
"loss": 0.014,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.023326335432703522,
|
| 34 |
+
"grad_norm": 0.12974420189857483,
|
| 35 |
+
"learning_rate": 9.881991975454331e-05,
|
| 36 |
+
"loss": 0.0099,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0291579192908794,
|
| 41 |
+
"grad_norm": 0.16743576526641846,
|
| 42 |
+
"learning_rate": 9.822987963181498e-05,
|
| 43 |
+
"loss": 0.0101,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.03498950314905528,
|
| 48 |
+
"grad_norm": 0.2192072570323944,
|
| 49 |
+
"learning_rate": 9.763983950908663e-05,
|
| 50 |
+
"loss": 0.0095,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.040821087007231166,
|
| 55 |
+
"grad_norm": 0.1741335690021515,
|
| 56 |
+
"learning_rate": 9.704979938635828e-05,
|
| 57 |
+
"loss": 0.0096,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.046652670865407045,
|
| 62 |
+
"grad_norm": 0.11039887368679047,
|
| 63 |
+
"learning_rate": 9.645975926362993e-05,
|
| 64 |
+
"loss": 0.0072,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.052484254723582924,
|
| 69 |
+
"grad_norm": 0.16088618338108063,
|
| 70 |
+
"learning_rate": 9.586971914090159e-05,
|
| 71 |
+
"loss": 0.009,
|
| 72 |
+
"step": 225
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0583158385817588,
|
| 76 |
+
"grad_norm": 0.07560613751411438,
|
| 77 |
+
"learning_rate": 9.527967901817324e-05,
|
| 78 |
+
"loss": 0.0074,
|
| 79 |
+
"step": 250
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.06414742243993468,
|
| 83 |
+
"grad_norm": 0.0758112445473671,
|
| 84 |
+
"learning_rate": 9.46896388954449e-05,
|
| 85 |
+
"loss": 0.0092,
|
| 86 |
+
"step": 275
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.06997900629811056,
|
| 90 |
+
"grad_norm": 0.10088574141263962,
|
| 91 |
+
"learning_rate": 9.409959877271654e-05,
|
| 92 |
+
"loss": 0.008,
|
| 93 |
+
"step": 300
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.07581059015628645,
|
| 97 |
+
"grad_norm": 0.09721095860004425,
|
| 98 |
+
"learning_rate": 9.350955864998821e-05,
|
| 99 |
+
"loss": 0.0072,
|
| 100 |
+
"step": 325
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.08164217401446233,
|
| 104 |
+
"grad_norm": 0.11294808238744736,
|
| 105 |
+
"learning_rate": 9.291951852725986e-05,
|
| 106 |
+
"loss": 0.0069,
|
| 107 |
+
"step": 350
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.08747375787263821,
|
| 111 |
+
"grad_norm": 0.1476767361164093,
|
| 112 |
+
"learning_rate": 9.232947840453151e-05,
|
| 113 |
+
"loss": 0.009,
|
| 114 |
+
"step": 375
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.09330534173081409,
|
| 118 |
+
"grad_norm": 0.11318453401327133,
|
| 119 |
+
"learning_rate": 9.173943828180317e-05,
|
| 120 |
+
"loss": 0.0081,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.09913692558898997,
|
| 125 |
+
"grad_norm": 0.10059420764446259,
|
| 126 |
+
"learning_rate": 9.114939815907482e-05,
|
| 127 |
+
"loss": 0.0073,
|
| 128 |
+
"step": 425
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.10496850944716585,
|
| 132 |
+
"grad_norm": 0.0688878670334816,
|
| 133 |
+
"learning_rate": 9.055935803634647e-05,
|
| 134 |
+
"loss": 0.0071,
|
| 135 |
+
"step": 450
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.11080009330534173,
|
| 139 |
+
"grad_norm": 0.2254699021577835,
|
| 140 |
+
"learning_rate": 8.996931791361814e-05,
|
| 141 |
+
"loss": 0.0084,
|
| 142 |
+
"step": 475
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.1166316771635176,
|
| 146 |
+
"grad_norm": 0.09738397598266602,
|
| 147 |
+
"learning_rate": 8.937927779088978e-05,
|
| 148 |
+
"loss": 0.0074,
|
| 149 |
+
"step": 500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.1224632610216935,
|
| 153 |
+
"grad_norm": 0.10569097846746445,
|
| 154 |
+
"learning_rate": 8.878923766816144e-05,
|
| 155 |
+
"loss": 0.0076,
|
| 156 |
+
"step": 525
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.12829484487986936,
|
| 160 |
+
"grad_norm": 0.09912018477916718,
|
| 161 |
+
"learning_rate": 8.81991975454331e-05,
|
| 162 |
+
"loss": 0.0082,
|
| 163 |
+
"step": 550
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.13412642873804526,
|
| 167 |
+
"grad_norm": 0.06858157366514206,
|
| 168 |
+
"learning_rate": 8.760915742270475e-05,
|
| 169 |
+
"loss": 0.008,
|
| 170 |
+
"step": 575
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.13995801259622112,
|
| 174 |
+
"grad_norm": 0.09230218082666397,
|
| 175 |
+
"learning_rate": 8.70191172999764e-05,
|
| 176 |
+
"loss": 0.008,
|
| 177 |
+
"step": 600
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.145789596454397,
|
| 181 |
+
"grad_norm": 0.09949750453233719,
|
| 182 |
+
"learning_rate": 8.642907717724805e-05,
|
| 183 |
+
"loss": 0.008,
|
| 184 |
+
"step": 625
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.1516211803125729,
|
| 188 |
+
"grad_norm": 0.1029815599322319,
|
| 189 |
+
"learning_rate": 8.58390370545197e-05,
|
| 190 |
+
"loss": 0.008,
|
| 191 |
+
"step": 650
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.15745276417074877,
|
| 195 |
+
"grad_norm": 0.05565624684095383,
|
| 196 |
+
"learning_rate": 8.524899693179137e-05,
|
| 197 |
+
"loss": 0.0076,
|
| 198 |
+
"step": 675
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.16328434802892466,
|
| 202 |
+
"grad_norm": 0.07810457050800323,
|
| 203 |
+
"learning_rate": 8.465895680906301e-05,
|
| 204 |
+
"loss": 0.0082,
|
| 205 |
+
"step": 700
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.16911593188710053,
|
| 209 |
+
"grad_norm": 0.16746339201927185,
|
| 210 |
+
"learning_rate": 8.406891668633468e-05,
|
| 211 |
+
"loss": 0.0076,
|
| 212 |
+
"step": 725
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.17494751574527642,
|
| 216 |
+
"grad_norm": 0.2016323357820511,
|
| 217 |
+
"learning_rate": 8.347887656360633e-05,
|
| 218 |
+
"loss": 0.007,
|
| 219 |
+
"step": 750
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1807790996034523,
|
| 223 |
+
"grad_norm": 0.11883407831192017,
|
| 224 |
+
"learning_rate": 8.288883644087798e-05,
|
| 225 |
+
"loss": 0.0074,
|
| 226 |
+
"step": 775
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.18661068346162818,
|
| 230 |
+
"grad_norm": 0.08053091168403625,
|
| 231 |
+
"learning_rate": 8.229879631814964e-05,
|
| 232 |
+
"loss": 0.0079,
|
| 233 |
+
"step": 800
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.19244226731980407,
|
| 237 |
+
"grad_norm": 0.05473201349377632,
|
| 238 |
+
"learning_rate": 8.170875619542129e-05,
|
| 239 |
+
"loss": 0.0072,
|
| 240 |
+
"step": 825
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.19827385117797994,
|
| 244 |
+
"grad_norm": 0.06678241491317749,
|
| 245 |
+
"learning_rate": 8.111871607269294e-05,
|
| 246 |
+
"loss": 0.0068,
|
| 247 |
+
"step": 850
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.20410543503615583,
|
| 251 |
+
"grad_norm": 0.04147233068943024,
|
| 252 |
+
"learning_rate": 8.05286759499646e-05,
|
| 253 |
+
"loss": 0.0071,
|
| 254 |
+
"step": 875
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.2099370188943317,
|
| 258 |
+
"grad_norm": 0.14302900433540344,
|
| 259 |
+
"learning_rate": 7.993863582723626e-05,
|
| 260 |
+
"loss": 0.0069,
|
| 261 |
+
"step": 900
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.2157686027525076,
|
| 265 |
+
"grad_norm": 0.07474330812692642,
|
| 266 |
+
"learning_rate": 7.934859570450791e-05,
|
| 267 |
+
"loss": 0.0081,
|
| 268 |
+
"step": 925
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.22160018661068345,
|
| 272 |
+
"grad_norm": 0.053421154618263245,
|
| 273 |
+
"learning_rate": 7.875855558177956e-05,
|
| 274 |
+
"loss": 0.0074,
|
| 275 |
+
"step": 950
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.22743177046885935,
|
| 279 |
+
"grad_norm": 0.10754521191120148,
|
| 280 |
+
"learning_rate": 7.816851545905122e-05,
|
| 281 |
+
"loss": 0.0068,
|
| 282 |
+
"step": 975
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.2332633543270352,
|
| 286 |
+
"grad_norm": 0.028845086693763733,
|
| 287 |
+
"learning_rate": 7.757847533632287e-05,
|
| 288 |
+
"loss": 0.0064,
|
| 289 |
+
"step": 1000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.2390949381852111,
|
| 293 |
+
"grad_norm": 0.08012504875659943,
|
| 294 |
+
"learning_rate": 7.698843521359452e-05,
|
| 295 |
+
"loss": 0.007,
|
| 296 |
+
"step": 1025
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.244926522043387,
|
| 300 |
+
"grad_norm": 0.12295565754175186,
|
| 301 |
+
"learning_rate": 7.639839509086619e-05,
|
| 302 |
+
"loss": 0.0064,
|
| 303 |
+
"step": 1050
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.2507581059015629,
|
| 307 |
+
"grad_norm": 0.07045774906873703,
|
| 308 |
+
"learning_rate": 7.580835496813784e-05,
|
| 309 |
+
"loss": 0.0077,
|
| 310 |
+
"step": 1075
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.2565896897597387,
|
| 314 |
+
"grad_norm": 0.07725486904382706,
|
| 315 |
+
"learning_rate": 7.521831484540949e-05,
|
| 316 |
+
"loss": 0.0058,
|
| 317 |
+
"step": 1100
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.2624212736179146,
|
| 321 |
+
"grad_norm": 0.062186215072870255,
|
| 322 |
+
"learning_rate": 7.462827472268115e-05,
|
| 323 |
+
"loss": 0.0082,
|
| 324 |
+
"step": 1125
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.2682528574760905,
|
| 328 |
+
"grad_norm": 0.04699672758579254,
|
| 329 |
+
"learning_rate": 7.403823459995281e-05,
|
| 330 |
+
"loss": 0.0064,
|
| 331 |
+
"step": 1150
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.2740844413342664,
|
| 335 |
+
"grad_norm": 0.0840052142739296,
|
| 336 |
+
"learning_rate": 7.344819447722445e-05,
|
| 337 |
+
"loss": 0.0083,
|
| 338 |
+
"step": 1175
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.27991602519244224,
|
| 342 |
+
"grad_norm": 0.08768948912620544,
|
| 343 |
+
"learning_rate": 7.285815435449612e-05,
|
| 344 |
+
"loss": 0.0067,
|
| 345 |
+
"step": 1200
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.28574760905061813,
|
| 349 |
+
"grad_norm": 0.11442771553993225,
|
| 350 |
+
"learning_rate": 7.226811423176776e-05,
|
| 351 |
+
"loss": 0.0068,
|
| 352 |
+
"step": 1225
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.291579192908794,
|
| 356 |
+
"grad_norm": 0.04983456805348396,
|
| 357 |
+
"learning_rate": 7.167807410903942e-05,
|
| 358 |
+
"loss": 0.0079,
|
| 359 |
+
"step": 1250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.2974107767669699,
|
| 363 |
+
"grad_norm": 0.0609428770840168,
|
| 364 |
+
"learning_rate": 7.108803398631107e-05,
|
| 365 |
+
"loss": 0.0074,
|
| 366 |
+
"step": 1275
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.3032423606251458,
|
| 370 |
+
"grad_norm": 0.04944339394569397,
|
| 371 |
+
"learning_rate": 7.049799386358273e-05,
|
| 372 |
+
"loss": 0.008,
|
| 373 |
+
"step": 1300
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.30907394448332165,
|
| 377 |
+
"grad_norm": 0.07327156513929367,
|
| 378 |
+
"learning_rate": 6.990795374085438e-05,
|
| 379 |
+
"loss": 0.0079,
|
| 380 |
+
"step": 1325
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.31490552834149754,
|
| 384 |
+
"grad_norm": 0.04584205895662308,
|
| 385 |
+
"learning_rate": 6.931791361812605e-05,
|
| 386 |
+
"loss": 0.0084,
|
| 387 |
+
"step": 1350
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.32073711219967344,
|
| 391 |
+
"grad_norm": 0.02472819946706295,
|
| 392 |
+
"learning_rate": 6.872787349539768e-05,
|
| 393 |
+
"loss": 0.0068,
|
| 394 |
+
"step": 1375
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.32656869605784933,
|
| 398 |
+
"grad_norm": 0.04464813694357872,
|
| 399 |
+
"learning_rate": 6.813783337266935e-05,
|
| 400 |
+
"loss": 0.0068,
|
| 401 |
+
"step": 1400
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.33240027991602517,
|
| 405 |
+
"grad_norm": 0.1550447791814804,
|
| 406 |
+
"learning_rate": 6.754779324994099e-05,
|
| 407 |
+
"loss": 0.0085,
|
| 408 |
+
"step": 1425
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.33823186377420106,
|
| 412 |
+
"grad_norm": 0.07317336648702621,
|
| 413 |
+
"learning_rate": 6.695775312721266e-05,
|
| 414 |
+
"loss": 0.0078,
|
| 415 |
+
"step": 1450
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.34406344763237695,
|
| 419 |
+
"grad_norm": 0.08646785467863083,
|
| 420 |
+
"learning_rate": 6.636771300448431e-05,
|
| 421 |
+
"loss": 0.0089,
|
| 422 |
+
"step": 1475
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.34989503149055284,
|
| 426 |
+
"grad_norm": 0.05935555696487427,
|
| 427 |
+
"learning_rate": 6.577767288175596e-05,
|
| 428 |
+
"loss": 0.0079,
|
| 429 |
+
"step": 1500
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.35572661534872874,
|
| 433 |
+
"grad_norm": 0.045770954340696335,
|
| 434 |
+
"learning_rate": 6.518763275902761e-05,
|
| 435 |
+
"loss": 0.0075,
|
| 436 |
+
"step": 1525
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.3615581992069046,
|
| 440 |
+
"grad_norm": 0.060902174562215805,
|
| 441 |
+
"learning_rate": 6.459759263629928e-05,
|
| 442 |
+
"loss": 0.0063,
|
| 443 |
+
"step": 1550
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.36738978306508047,
|
| 447 |
+
"grad_norm": 0.11909040063619614,
|
| 448 |
+
"learning_rate": 6.400755251357092e-05,
|
| 449 |
+
"loss": 0.0068,
|
| 450 |
+
"step": 1575
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.37322136692325636,
|
| 454 |
+
"grad_norm": 0.10495050251483917,
|
| 455 |
+
"learning_rate": 6.341751239084258e-05,
|
| 456 |
+
"loss": 0.0074,
|
| 457 |
+
"step": 1600
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.37905295078143225,
|
| 461 |
+
"grad_norm": 0.08483859896659851,
|
| 462 |
+
"learning_rate": 6.282747226811424e-05,
|
| 463 |
+
"loss": 0.0066,
|
| 464 |
+
"step": 1625
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.38488453463960814,
|
| 468 |
+
"grad_norm": 0.05936511233448982,
|
| 469 |
+
"learning_rate": 6.223743214538589e-05,
|
| 470 |
+
"loss": 0.0068,
|
| 471 |
+
"step": 1650
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.390716118497784,
|
| 475 |
+
"grad_norm": 0.03783701732754707,
|
| 476 |
+
"learning_rate": 6.164739202265754e-05,
|
| 477 |
+
"loss": 0.0067,
|
| 478 |
+
"step": 1675
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.3965477023559599,
|
| 482 |
+
"grad_norm": 0.03322950750589371,
|
| 483 |
+
"learning_rate": 6.10573518999292e-05,
|
| 484 |
+
"loss": 0.0067,
|
| 485 |
+
"step": 1700
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.40237928621413577,
|
| 489 |
+
"grad_norm": 0.06542570888996124,
|
| 490 |
+
"learning_rate": 6.0467311777200854e-05,
|
| 491 |
+
"loss": 0.007,
|
| 492 |
+
"step": 1725
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.40821087007231166,
|
| 496 |
+
"grad_norm": 0.05515117570757866,
|
| 497 |
+
"learning_rate": 5.987727165447251e-05,
|
| 498 |
+
"loss": 0.0069,
|
| 499 |
+
"step": 1750
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.4140424539304875,
|
| 503 |
+
"grad_norm": 0.06597180664539337,
|
| 504 |
+
"learning_rate": 5.928723153174416e-05,
|
| 505 |
+
"loss": 0.0066,
|
| 506 |
+
"step": 1775
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.4198740377886634,
|
| 510 |
+
"grad_norm": 0.14175090193748474,
|
| 511 |
+
"learning_rate": 5.869719140901582e-05,
|
| 512 |
+
"loss": 0.0072,
|
| 513 |
+
"step": 1800
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.4257056216468393,
|
| 517 |
+
"grad_norm": 0.08393794298171997,
|
| 518 |
+
"learning_rate": 5.810715128628748e-05,
|
| 519 |
+
"loss": 0.007,
|
| 520 |
+
"step": 1825
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.4315372055050152,
|
| 524 |
+
"grad_norm": 0.05573498085141182,
|
| 525 |
+
"learning_rate": 5.751711116355912e-05,
|
| 526 |
+
"loss": 0.0079,
|
| 527 |
+
"step": 1850
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.43736878936319107,
|
| 531 |
+
"grad_norm": 0.08545993268489838,
|
| 532 |
+
"learning_rate": 5.692707104083078e-05,
|
| 533 |
+
"loss": 0.0073,
|
| 534 |
+
"step": 1875
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.4432003732213669,
|
| 538 |
+
"grad_norm": 0.05015942454338074,
|
| 539 |
+
"learning_rate": 5.633703091810243e-05,
|
| 540 |
+
"loss": 0.0078,
|
| 541 |
+
"step": 1900
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.4490319570795428,
|
| 545 |
+
"grad_norm": 0.08075055480003357,
|
| 546 |
+
"learning_rate": 5.574699079537409e-05,
|
| 547 |
+
"loss": 0.0078,
|
| 548 |
+
"step": 1925
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.4548635409377187,
|
| 552 |
+
"grad_norm": 0.027930203825235367,
|
| 553 |
+
"learning_rate": 5.515695067264575e-05,
|
| 554 |
+
"loss": 0.0068,
|
| 555 |
+
"step": 1950
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.4606951247958946,
|
| 559 |
+
"grad_norm": 0.12208261340856552,
|
| 560 |
+
"learning_rate": 5.456691054991739e-05,
|
| 561 |
+
"loss": 0.0076,
|
| 562 |
+
"step": 1975
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.4665267086540704,
|
| 566 |
+
"grad_norm": 0.07081254571676254,
|
| 567 |
+
"learning_rate": 5.397687042718905e-05,
|
| 568 |
+
"loss": 0.0073,
|
| 569 |
+
"step": 2000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.4723582925122463,
|
| 573 |
+
"grad_norm": 0.06605388969182968,
|
| 574 |
+
"learning_rate": 5.338683030446071e-05,
|
| 575 |
+
"loss": 0.0067,
|
| 576 |
+
"step": 2025
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.4781898763704222,
|
| 580 |
+
"grad_norm": 0.07699355483055115,
|
| 581 |
+
"learning_rate": 5.279679018173236e-05,
|
| 582 |
+
"loss": 0.0074,
|
| 583 |
+
"step": 2050
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.4840214602285981,
|
| 587 |
+
"grad_norm": 0.07484355568885803,
|
| 588 |
+
"learning_rate": 5.2206750059004016e-05,
|
| 589 |
+
"loss": 0.008,
|
| 590 |
+
"step": 2075
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.489853044086774,
|
| 594 |
+
"grad_norm": 0.05196300894021988,
|
| 595 |
+
"learning_rate": 5.161670993627566e-05,
|
| 596 |
+
"loss": 0.0086,
|
| 597 |
+
"step": 2100
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.49568462794494983,
|
| 601 |
+
"grad_norm": 0.07191181927919388,
|
| 602 |
+
"learning_rate": 5.102666981354732e-05,
|
| 603 |
+
"loss": 0.0063,
|
| 604 |
+
"step": 2125
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.5015162118031258,
|
| 608 |
+
"grad_norm": 0.12823079526424408,
|
| 609 |
+
"learning_rate": 5.043662969081898e-05,
|
| 610 |
+
"loss": 0.0065,
|
| 611 |
+
"step": 2150
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.5073477956613016,
|
| 615 |
+
"grad_norm": 0.05935099348425865,
|
| 616 |
+
"learning_rate": 4.9846589568090633e-05,
|
| 617 |
+
"loss": 0.0072,
|
| 618 |
+
"step": 2175
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.5131793795194775,
|
| 622 |
+
"grad_norm": 0.07369107007980347,
|
| 623 |
+
"learning_rate": 4.9256549445362286e-05,
|
| 624 |
+
"loss": 0.0071,
|
| 625 |
+
"step": 2200
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.5190109633776534,
|
| 629 |
+
"grad_norm": 0.05355490744113922,
|
| 630 |
+
"learning_rate": 4.866650932263394e-05,
|
| 631 |
+
"loss": 0.007,
|
| 632 |
+
"step": 2225
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.5248425472358292,
|
| 636 |
+
"grad_norm": 0.0854572206735611,
|
| 637 |
+
"learning_rate": 4.807646919990559e-05,
|
| 638 |
+
"loss": 0.0069,
|
| 639 |
+
"step": 2250
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.5306741310940052,
|
| 643 |
+
"grad_norm": 0.058111775666475296,
|
| 644 |
+
"learning_rate": 4.748642907717725e-05,
|
| 645 |
+
"loss": 0.0074,
|
| 646 |
+
"step": 2275
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.536505714952181,
|
| 650 |
+
"grad_norm": 0.06656444817781448,
|
| 651 |
+
"learning_rate": 4.68963889544489e-05,
|
| 652 |
+
"loss": 0.0064,
|
| 653 |
+
"step": 2300
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.5423372988103569,
|
| 657 |
+
"grad_norm": 0.08224783837795258,
|
| 658 |
+
"learning_rate": 4.6306348831720555e-05,
|
| 659 |
+
"loss": 0.0082,
|
| 660 |
+
"step": 2325
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.5481688826685328,
|
| 664 |
+
"grad_norm": 0.06290777772665024,
|
| 665 |
+
"learning_rate": 4.5716308708992215e-05,
|
| 666 |
+
"loss": 0.0066,
|
| 667 |
+
"step": 2350
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.5540004665267086,
|
| 671 |
+
"grad_norm": 0.046868737787008286,
|
| 672 |
+
"learning_rate": 4.512626858626387e-05,
|
| 673 |
+
"loss": 0.0064,
|
| 674 |
+
"step": 2375
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.5598320503848845,
|
| 678 |
+
"grad_norm": 0.08161591738462448,
|
| 679 |
+
"learning_rate": 4.453622846353552e-05,
|
| 680 |
+
"loss": 0.0086,
|
| 681 |
+
"step": 2400
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.5656636342430604,
|
| 685 |
+
"grad_norm": 0.06194046139717102,
|
| 686 |
+
"learning_rate": 4.394618834080718e-05,
|
| 687 |
+
"loss": 0.0071,
|
| 688 |
+
"step": 2425
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.5714952181012363,
|
| 692 |
+
"grad_norm": 0.06799814105033875,
|
| 693 |
+
"learning_rate": 4.335614821807883e-05,
|
| 694 |
+
"loss": 0.0065,
|
| 695 |
+
"step": 2450
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.5773268019594122,
|
| 699 |
+
"grad_norm": 0.06985451281070709,
|
| 700 |
+
"learning_rate": 4.276610809535049e-05,
|
| 701 |
+
"loss": 0.0096,
|
| 702 |
+
"step": 2475
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.583158385817588,
|
| 706 |
+
"grad_norm": 0.045887541025877,
|
| 707 |
+
"learning_rate": 4.2176067972622144e-05,
|
| 708 |
+
"loss": 0.0066,
|
| 709 |
+
"step": 2500
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.5889899696757639,
|
| 713 |
+
"grad_norm": 0.10378813743591309,
|
| 714 |
+
"learning_rate": 4.1586027849893796e-05,
|
| 715 |
+
"loss": 0.0066,
|
| 716 |
+
"step": 2525
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.5948215535339398,
|
| 720 |
+
"grad_norm": 0.037475138902664185,
|
| 721 |
+
"learning_rate": 4.099598772716545e-05,
|
| 722 |
+
"loss": 0.0077,
|
| 723 |
+
"step": 2550
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.6006531373921157,
|
| 727 |
+
"grad_norm": 0.13552173972129822,
|
| 728 |
+
"learning_rate": 4.040594760443711e-05,
|
| 729 |
+
"loss": 0.0082,
|
| 730 |
+
"step": 2575
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.6064847212502916,
|
| 734 |
+
"grad_norm": 0.049305260181427,
|
| 735 |
+
"learning_rate": 3.981590748170876e-05,
|
| 736 |
+
"loss": 0.0063,
|
| 737 |
+
"step": 2600
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.6123163051084675,
|
| 741 |
+
"grad_norm": 0.0553220734000206,
|
| 742 |
+
"learning_rate": 3.922586735898041e-05,
|
| 743 |
+
"loss": 0.0079,
|
| 744 |
+
"step": 2625
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.6181478889666433,
|
| 748 |
+
"grad_norm": 0.043462011963129044,
|
| 749 |
+
"learning_rate": 3.8635827236252066e-05,
|
| 750 |
+
"loss": 0.0067,
|
| 751 |
+
"step": 2650
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.6239794728248192,
|
| 755 |
+
"grad_norm": 0.331167995929718,
|
| 756 |
+
"learning_rate": 3.8045787113523725e-05,
|
| 757 |
+
"loss": 0.008,
|
| 758 |
+
"step": 2675
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.6298110566829951,
|
| 762 |
+
"grad_norm": 0.06077446788549423,
|
| 763 |
+
"learning_rate": 3.745574699079538e-05,
|
| 764 |
+
"loss": 0.0067,
|
| 765 |
+
"step": 2700
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.635642640541171,
|
| 769 |
+
"grad_norm": 0.06422346830368042,
|
| 770 |
+
"learning_rate": 3.686570686806703e-05,
|
| 771 |
+
"loss": 0.0072,
|
| 772 |
+
"step": 2725
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.6414742243993469,
|
| 776 |
+
"grad_norm": 0.053520068526268005,
|
| 777 |
+
"learning_rate": 3.627566674533868e-05,
|
| 778 |
+
"loss": 0.0057,
|
| 779 |
+
"step": 2750
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.6473058082575227,
|
| 783 |
+
"grad_norm": 0.08369456231594086,
|
| 784 |
+
"learning_rate": 3.568562662261034e-05,
|
| 785 |
+
"loss": 0.0071,
|
| 786 |
+
"step": 2775
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.6531373921156987,
|
| 790 |
+
"grad_norm": 0.07306842505931854,
|
| 791 |
+
"learning_rate": 3.5095586499881994e-05,
|
| 792 |
+
"loss": 0.0065,
|
| 793 |
+
"step": 2800
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.6589689759738745,
|
| 797 |
+
"grad_norm": 0.04199352487921715,
|
| 798 |
+
"learning_rate": 3.450554637715365e-05,
|
| 799 |
+
"loss": 0.007,
|
| 800 |
+
"step": 2825
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.6648005598320503,
|
| 804 |
+
"grad_norm": 0.04420465603470802,
|
| 805 |
+
"learning_rate": 3.39155062544253e-05,
|
| 806 |
+
"loss": 0.0064,
|
| 807 |
+
"step": 2850
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.6706321436902263,
|
| 811 |
+
"grad_norm": 0.09053666144609451,
|
| 812 |
+
"learning_rate": 3.332546613169696e-05,
|
| 813 |
+
"loss": 0.0057,
|
| 814 |
+
"step": 2875
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.6764637275484021,
|
| 818 |
+
"grad_norm": 0.04140497371554375,
|
| 819 |
+
"learning_rate": 3.273542600896861e-05,
|
| 820 |
+
"loss": 0.0067,
|
| 821 |
+
"step": 2900
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.6822953114065781,
|
| 825 |
+
"grad_norm": 0.038969893008470535,
|
| 826 |
+
"learning_rate": 3.2145385886240264e-05,
|
| 827 |
+
"loss": 0.0059,
|
| 828 |
+
"step": 2925
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.6881268952647539,
|
| 832 |
+
"grad_norm": 0.05273483693599701,
|
| 833 |
+
"learning_rate": 3.1555345763511916e-05,
|
| 834 |
+
"loss": 0.0065,
|
| 835 |
+
"step": 2950
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.6939584791229297,
|
| 839 |
+
"grad_norm": 0.0726546123623848,
|
| 840 |
+
"learning_rate": 3.0965305640783576e-05,
|
| 841 |
+
"loss": 0.0061,
|
| 842 |
+
"step": 2975
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.6997900629811057,
|
| 846 |
+
"grad_norm": 0.07580599933862686,
|
| 847 |
+
"learning_rate": 3.0375265518055228e-05,
|
| 848 |
+
"loss": 0.0068,
|
| 849 |
+
"step": 3000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.7056216468392815,
|
| 853 |
+
"grad_norm": 0.036154165863990784,
|
| 854 |
+
"learning_rate": 2.978522539532688e-05,
|
| 855 |
+
"loss": 0.0078,
|
| 856 |
+
"step": 3025
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.7114532306974575,
|
| 860 |
+
"grad_norm": 0.07397032529115677,
|
| 861 |
+
"learning_rate": 2.919518527259854e-05,
|
| 862 |
+
"loss": 0.006,
|
| 863 |
+
"step": 3050
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.7172848145556333,
|
| 867 |
+
"grad_norm": 0.11983916163444519,
|
| 868 |
+
"learning_rate": 2.8605145149870193e-05,
|
| 869 |
+
"loss": 0.0067,
|
| 870 |
+
"step": 3075
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.7231163984138091,
|
| 874 |
+
"grad_norm": 0.0615057498216629,
|
| 875 |
+
"learning_rate": 2.801510502714185e-05,
|
| 876 |
+
"loss": 0.0079,
|
| 877 |
+
"step": 3100
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.7289479822719851,
|
| 881 |
+
"grad_norm": 0.03690054267644882,
|
| 882 |
+
"learning_rate": 2.74250649044135e-05,
|
| 883 |
+
"loss": 0.0067,
|
| 884 |
+
"step": 3125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.7347795661301609,
|
| 888 |
+
"grad_norm": 0.07274270802736282,
|
| 889 |
+
"learning_rate": 2.6835024781685157e-05,
|
| 890 |
+
"loss": 0.0073,
|
| 891 |
+
"step": 3150
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.7406111499883369,
|
| 895 |
+
"grad_norm": 0.0663859024643898,
|
| 896 |
+
"learning_rate": 2.6244984658956813e-05,
|
| 897 |
+
"loss": 0.0071,
|
| 898 |
+
"step": 3175
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.7464427338465127,
|
| 902 |
+
"grad_norm": 0.05553248152136803,
|
| 903 |
+
"learning_rate": 2.5654944536228466e-05,
|
| 904 |
+
"loss": 0.0075,
|
| 905 |
+
"step": 3200
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.7522743177046886,
|
| 909 |
+
"grad_norm": 0.04285744950175285,
|
| 910 |
+
"learning_rate": 2.5064904413500118e-05,
|
| 911 |
+
"loss": 0.0066,
|
| 912 |
+
"step": 3225
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.7581059015628645,
|
| 916 |
+
"grad_norm": 0.09008394181728363,
|
| 917 |
+
"learning_rate": 2.4474864290771774e-05,
|
| 918 |
+
"loss": 0.0076,
|
| 919 |
+
"step": 3250
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.7639374854210403,
|
| 923 |
+
"grad_norm": 0.058279018849134445,
|
| 924 |
+
"learning_rate": 2.388482416804343e-05,
|
| 925 |
+
"loss": 0.0075,
|
| 926 |
+
"step": 3275
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.7697690692792163,
|
| 930 |
+
"grad_norm": 0.05219094827771187,
|
| 931 |
+
"learning_rate": 2.3294784045315082e-05,
|
| 932 |
+
"loss": 0.0071,
|
| 933 |
+
"step": 3300
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.7756006531373921,
|
| 937 |
+
"grad_norm": 0.05435522645711899,
|
| 938 |
+
"learning_rate": 2.270474392258674e-05,
|
| 939 |
+
"loss": 0.006,
|
| 940 |
+
"step": 3325
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.781432236995568,
|
| 944 |
+
"grad_norm": 0.05999380722641945,
|
| 945 |
+
"learning_rate": 2.211470379985839e-05,
|
| 946 |
+
"loss": 0.0084,
|
| 947 |
+
"step": 3350
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.7872638208537439,
|
| 951 |
+
"grad_norm": 0.029552578926086426,
|
| 952 |
+
"learning_rate": 2.1524663677130047e-05,
|
| 953 |
+
"loss": 0.0061,
|
| 954 |
+
"step": 3375
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.7930954047119197,
|
| 958 |
+
"grad_norm": 0.039268866181373596,
|
| 959 |
+
"learning_rate": 2.09346235544017e-05,
|
| 960 |
+
"loss": 0.0069,
|
| 961 |
+
"step": 3400
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.7989269885700956,
|
| 965 |
+
"grad_norm": 0.05782666802406311,
|
| 966 |
+
"learning_rate": 2.0344583431673355e-05,
|
| 967 |
+
"loss": 0.0067,
|
| 968 |
+
"step": 3425
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.8047585724282715,
|
| 972 |
+
"grad_norm": 0.03869437053799629,
|
| 973 |
+
"learning_rate": 1.9754543308945008e-05,
|
| 974 |
+
"loss": 0.0066,
|
| 975 |
+
"step": 3450
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.8105901562864474,
|
| 979 |
+
"grad_norm": 0.05132381245493889,
|
| 980 |
+
"learning_rate": 1.9164503186216664e-05,
|
| 981 |
+
"loss": 0.0064,
|
| 982 |
+
"step": 3475
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.8164217401446233,
|
| 986 |
+
"grad_norm": 0.08721393346786499,
|
| 987 |
+
"learning_rate": 1.8574463063488316e-05,
|
| 988 |
+
"loss": 0.0084,
|
| 989 |
+
"step": 3500
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.8222533240027992,
|
| 993 |
+
"grad_norm": 0.08003130555152893,
|
| 994 |
+
"learning_rate": 1.7984422940759972e-05,
|
| 995 |
+
"loss": 0.0058,
|
| 996 |
+
"step": 3525
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.828084907860975,
|
| 1000 |
+
"grad_norm": 0.05190620571374893,
|
| 1001 |
+
"learning_rate": 1.7394382818031625e-05,
|
| 1002 |
+
"loss": 0.0066,
|
| 1003 |
+
"step": 3550
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.8339164917191509,
|
| 1007 |
+
"grad_norm": 0.06803402304649353,
|
| 1008 |
+
"learning_rate": 1.680434269530328e-05,
|
| 1009 |
+
"loss": 0.0069,
|
| 1010 |
+
"step": 3575
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.8397480755773268,
|
| 1014 |
+
"grad_norm": 0.05517745763063431,
|
| 1015 |
+
"learning_rate": 1.6214302572574937e-05,
|
| 1016 |
+
"loss": 0.0086,
|
| 1017 |
+
"step": 3600
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.8455796594355027,
|
| 1021 |
+
"grad_norm": 0.05708625167608261,
|
| 1022 |
+
"learning_rate": 1.5624262449846593e-05,
|
| 1023 |
+
"loss": 0.0063,
|
| 1024 |
+
"step": 3625
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.8514112432936786,
|
| 1028 |
+
"grad_norm": 0.05021214112639427,
|
| 1029 |
+
"learning_rate": 1.5034222327118243e-05,
|
| 1030 |
+
"loss": 0.0056,
|
| 1031 |
+
"step": 3650
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.8572428271518544,
|
| 1035 |
+
"grad_norm": 0.06552550941705704,
|
| 1036 |
+
"learning_rate": 1.44441822043899e-05,
|
| 1037 |
+
"loss": 0.0065,
|
| 1038 |
+
"step": 3675
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.8630744110100304,
|
| 1042 |
+
"grad_norm": 0.05650312826037407,
|
| 1043 |
+
"learning_rate": 1.3854142081661554e-05,
|
| 1044 |
+
"loss": 0.0071,
|
| 1045 |
+
"step": 3700
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.8689059948682062,
|
| 1049 |
+
"grad_norm": 0.021646451205015182,
|
| 1050 |
+
"learning_rate": 1.326410195893321e-05,
|
| 1051 |
+
"loss": 0.0056,
|
| 1052 |
+
"step": 3725
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.8747375787263821,
|
| 1056 |
+
"grad_norm": 0.04382350668311119,
|
| 1057 |
+
"learning_rate": 1.2674061836204862e-05,
|
| 1058 |
+
"loss": 0.0068,
|
| 1059 |
+
"step": 3750
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.880569162584558,
|
| 1063 |
+
"grad_norm": 0.06425108760595322,
|
| 1064 |
+
"learning_rate": 1.2084021713476516e-05,
|
| 1065 |
+
"loss": 0.008,
|
| 1066 |
+
"step": 3775
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.8864007464427338,
|
| 1070 |
+
"grad_norm": 0.07273637503385544,
|
| 1071 |
+
"learning_rate": 1.1493981590748172e-05,
|
| 1072 |
+
"loss": 0.0065,
|
| 1073 |
+
"step": 3800
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.8922323303009098,
|
| 1077 |
+
"grad_norm": 0.0765676200389862,
|
| 1078 |
+
"learning_rate": 1.0903941468019826e-05,
|
| 1079 |
+
"loss": 0.006,
|
| 1080 |
+
"step": 3825
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.8980639141590856,
|
| 1084 |
+
"grad_norm": 0.05742628127336502,
|
| 1085 |
+
"learning_rate": 1.031390134529148e-05,
|
| 1086 |
+
"loss": 0.007,
|
| 1087 |
+
"step": 3850
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.9038954980172614,
|
| 1091 |
+
"grad_norm": 0.059533748775720596,
|
| 1092 |
+
"learning_rate": 9.723861222563135e-06,
|
| 1093 |
+
"loss": 0.0064,
|
| 1094 |
+
"step": 3875
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.9097270818754374,
|
| 1098 |
+
"grad_norm": 0.02601277269423008,
|
| 1099 |
+
"learning_rate": 9.13382109983479e-06,
|
| 1100 |
+
"loss": 0.0061,
|
| 1101 |
+
"step": 3900
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.9155586657336132,
|
| 1105 |
+
"grad_norm": 0.04568032547831535,
|
| 1106 |
+
"learning_rate": 8.543780977106443e-06,
|
| 1107 |
+
"loss": 0.0064,
|
| 1108 |
+
"step": 3925
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.9213902495917892,
|
| 1112 |
+
"grad_norm": 0.04665420576930046,
|
| 1113 |
+
"learning_rate": 7.9537408543781e-06,
|
| 1114 |
+
"loss": 0.0072,
|
| 1115 |
+
"step": 3950
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.927221833449965,
|
| 1119 |
+
"grad_norm": 0.04735620319843292,
|
| 1120 |
+
"learning_rate": 7.363700731649753e-06,
|
| 1121 |
+
"loss": 0.0072,
|
| 1122 |
+
"step": 3975
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.9330534173081408,
|
| 1126 |
+
"grad_norm": 0.09656576067209244,
|
| 1127 |
+
"learning_rate": 6.773660608921408e-06,
|
| 1128 |
+
"loss": 0.0068,
|
| 1129 |
+
"step": 4000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.9388850011663168,
|
| 1133 |
+
"grad_norm": 0.05522262305021286,
|
| 1134 |
+
"learning_rate": 6.183620486193061e-06,
|
| 1135 |
+
"loss": 0.0056,
|
| 1136 |
+
"step": 4025
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.9447165850244926,
|
| 1140 |
+
"grad_norm": 0.05257203057408333,
|
| 1141 |
+
"learning_rate": 5.5935803634647154e-06,
|
| 1142 |
+
"loss": 0.0056,
|
| 1143 |
+
"step": 4050
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.9505481688826686,
|
| 1147 |
+
"grad_norm": 0.04314320161938667,
|
| 1148 |
+
"learning_rate": 5.00354024073637e-06,
|
| 1149 |
+
"loss": 0.0063,
|
| 1150 |
+
"step": 4075
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.9563797527408444,
|
| 1154 |
+
"grad_norm": 0.04623764008283615,
|
| 1155 |
+
"learning_rate": 4.413500118008025e-06,
|
| 1156 |
+
"loss": 0.0067,
|
| 1157 |
+
"step": 4100
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.9622113365990203,
|
| 1161 |
+
"grad_norm": 0.05779367685317993,
|
| 1162 |
+
"learning_rate": 3.823459995279679e-06,
|
| 1163 |
+
"loss": 0.007,
|
| 1164 |
+
"step": 4125
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.9680429204571962,
|
| 1168 |
+
"grad_norm": 0.07973254472017288,
|
| 1169 |
+
"learning_rate": 3.2334198725513332e-06,
|
| 1170 |
+
"loss": 0.0068,
|
| 1171 |
+
"step": 4150
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.973874504315372,
|
| 1175 |
+
"grad_norm": 0.056475065648555756,
|
| 1176 |
+
"learning_rate": 2.643379749822988e-06,
|
| 1177 |
+
"loss": 0.0056,
|
| 1178 |
+
"step": 4175
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.979706088173548,
|
| 1182 |
+
"grad_norm": 0.05260898917913437,
|
| 1183 |
+
"learning_rate": 2.0533396270946426e-06,
|
| 1184 |
+
"loss": 0.0061,
|
| 1185 |
+
"step": 4200
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.9855376720317238,
|
| 1189 |
+
"grad_norm": 0.07043284177780151,
|
| 1190 |
+
"learning_rate": 1.463299504366297e-06,
|
| 1191 |
+
"loss": 0.0061,
|
| 1192 |
+
"step": 4225
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.9913692558898997,
|
| 1196 |
+
"grad_norm": 0.09899988770484924,
|
| 1197 |
+
"learning_rate": 8.732593816379515e-07,
|
| 1198 |
+
"loss": 0.0076,
|
| 1199 |
+
"step": 4250
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.9972008397480756,
|
| 1203 |
+
"grad_norm": 0.03969600424170494,
|
| 1204 |
+
"learning_rate": 2.832192589096059e-07,
|
| 1205 |
+
"loss": 0.0067,
|
| 1206 |
+
"step": 4275
|
| 1207 |
+
}
|
| 1208 |
+
],
|
| 1209 |
+
"logging_steps": 25,
|
| 1210 |
+
"max_steps": 4287,
|
| 1211 |
+
"num_input_tokens_seen": 0,
|
| 1212 |
+
"num_train_epochs": 1,
|
| 1213 |
+
"save_steps": 250,
|
| 1214 |
+
"stateful_callbacks": {
|
| 1215 |
+
"TrainerControl": {
|
| 1216 |
+
"args": {
|
| 1217 |
+
"should_epoch_stop": false,
|
| 1218 |
+
"should_evaluate": false,
|
| 1219 |
+
"should_log": false,
|
| 1220 |
+
"should_save": true,
|
| 1221 |
+
"should_training_stop": true
|
| 1222 |
+
},
|
| 1223 |
+
"attributes": {}
|
| 1224 |
+
}
|
| 1225 |
+
},
|
| 1226 |
+
"total_flos": 4.927104921604435e+17,
|
| 1227 |
+
"train_batch_size": 1,
|
| 1228 |
+
"trial_name": null,
|
| 1229 |
+
"trial_params": null
|
| 1230 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0137c40ddb51c5ff6609f5ff00cc1bec6ff64ad21c910b4f29d13f8f2d48feb
|
| 3 |
+
size 5176
|