Initial model upload
Browse filesSigned-off-by: Andreas Florath <[email protected]>
- adapter_config.json +33 -0
- adapter_model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +951 -0
- training_args.bin +3 -0
adapter_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "mistralai/Mistral-7B-v0.1",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": {},
|
| 12 |
+
"lora_alpha": 64,
|
| 13 |
+
"lora_dropout": 0.05,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 32,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": null,
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"v_proj",
|
| 23 |
+
"down_proj",
|
| 24 |
+
"k_proj",
|
| 25 |
+
"o_proj",
|
| 26 |
+
"gate_proj",
|
| 27 |
+
"q_proj",
|
| 28 |
+
"lm_head",
|
| 29 |
+
"up_proj"
|
| 30 |
+
],
|
| 31 |
+
"task_type": "CAUSAL_LM",
|
| 32 |
+
"use_rslora": false
|
| 33 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c7c602870e0663ef50844dde38d8f4db0e36ea06046ed86f55d2177ea7f8c3fa
|
| 3 |
+
size 864513616
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fcec2b93a4a78c6b971ba811dc42f0253c5192755742c0b03eb3b09014d5955e
|
| 3 |
+
size 170951516
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ed2738cc0ced287aa856f39260a877fb100a77a846f1bbc502c43c3c7b19e18b
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc42fadd0d8fe8dc7c0b2b8c0265ef1e2908ab518681e3c308a62f5221956d1a
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,951 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.835446953294154,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 12400,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.06,
|
| 13 |
+
"grad_norm": 1.6350187063217163,
|
| 14 |
+
"learning_rate": 4.977888397519945e-05,
|
| 15 |
+
"loss": 1.2168,
|
| 16 |
+
"step": 200
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.06,
|
| 20 |
+
"eval_loss": 1.0502910614013672,
|
| 21 |
+
"eval_runtime": 524.206,
|
| 22 |
+
"eval_samples_per_second": 12.058,
|
| 23 |
+
"eval_steps_per_second": 1.509,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.12,
|
| 28 |
+
"grad_norm": 1.5651005506515503,
|
| 29 |
+
"learning_rate": 4.955665681459588e-05,
|
| 30 |
+
"loss": 0.8535,
|
| 31 |
+
"step": 400
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.12,
|
| 35 |
+
"eval_loss": 0.9054814577102661,
|
| 36 |
+
"eval_runtime": 523.7675,
|
| 37 |
+
"eval_samples_per_second": 12.068,
|
| 38 |
+
"eval_steps_per_second": 1.51,
|
| 39 |
+
"step": 400
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.19,
|
| 43 |
+
"grad_norm": 1.4049625396728516,
|
| 44 |
+
"learning_rate": 4.9334429653992315e-05,
|
| 45 |
+
"loss": 0.7543,
|
| 46 |
+
"step": 600
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.19,
|
| 50 |
+
"eval_loss": 0.8259416818618774,
|
| 51 |
+
"eval_runtime": 524.2403,
|
| 52 |
+
"eval_samples_per_second": 12.057,
|
| 53 |
+
"eval_steps_per_second": 1.509,
|
| 54 |
+
"step": 600
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.25,
|
| 58 |
+
"grad_norm": 1.6197569370269775,
|
| 59 |
+
"learning_rate": 4.911220249338874e-05,
|
| 60 |
+
"loss": 0.7007,
|
| 61 |
+
"step": 800
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.25,
|
| 65 |
+
"eval_loss": 0.7792428135871887,
|
| 66 |
+
"eval_runtime": 523.8718,
|
| 67 |
+
"eval_samples_per_second": 12.066,
|
| 68 |
+
"eval_steps_per_second": 1.51,
|
| 69 |
+
"step": 800
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.31,
|
| 73 |
+
"grad_norm": 1.3332139253616333,
|
| 74 |
+
"learning_rate": 4.888997533278518e-05,
|
| 75 |
+
"loss": 0.653,
|
| 76 |
+
"step": 1000
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.31,
|
| 80 |
+
"eval_loss": 0.7383475303649902,
|
| 81 |
+
"eval_runtime": 524.1649,
|
| 82 |
+
"eval_samples_per_second": 12.059,
|
| 83 |
+
"eval_steps_per_second": 1.509,
|
| 84 |
+
"step": 1000
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.37,
|
| 88 |
+
"grad_norm": 1.3490664958953857,
|
| 89 |
+
"learning_rate": 4.8667748172181604e-05,
|
| 90 |
+
"loss": 0.6267,
|
| 91 |
+
"step": 1200
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.37,
|
| 95 |
+
"eval_loss": 0.7095275521278381,
|
| 96 |
+
"eval_runtime": 523.9699,
|
| 97 |
+
"eval_samples_per_second": 12.064,
|
| 98 |
+
"eval_steps_per_second": 1.51,
|
| 99 |
+
"step": 1200
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.43,
|
| 103 |
+
"grad_norm": 1.2756898403167725,
|
| 104 |
+
"learning_rate": 4.844552101157804e-05,
|
| 105 |
+
"loss": 0.5969,
|
| 106 |
+
"step": 1400
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.43,
|
| 110 |
+
"eval_loss": 0.6869618892669678,
|
| 111 |
+
"eval_runtime": 524.0331,
|
| 112 |
+
"eval_samples_per_second": 12.062,
|
| 113 |
+
"eval_steps_per_second": 1.509,
|
| 114 |
+
"step": 1400
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.49,
|
| 118 |
+
"grad_norm": 1.2612615823745728,
|
| 119 |
+
"learning_rate": 4.8223293850974466e-05,
|
| 120 |
+
"loss": 0.5771,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.49,
|
| 125 |
+
"eval_loss": 0.6662651300430298,
|
| 126 |
+
"eval_runtime": 524.2978,
|
| 127 |
+
"eval_samples_per_second": 12.056,
|
| 128 |
+
"eval_steps_per_second": 1.509,
|
| 129 |
+
"step": 1600
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.56,
|
| 133 |
+
"grad_norm": 1.1881535053253174,
|
| 134 |
+
"learning_rate": 4.80010666903709e-05,
|
| 135 |
+
"loss": 0.5632,
|
| 136 |
+
"step": 1800
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.56,
|
| 140 |
+
"eval_loss": 0.6456959247589111,
|
| 141 |
+
"eval_runtime": 524.0526,
|
| 142 |
+
"eval_samples_per_second": 12.062,
|
| 143 |
+
"eval_steps_per_second": 1.509,
|
| 144 |
+
"step": 1800
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.62,
|
| 148 |
+
"grad_norm": 1.1880247592926025,
|
| 149 |
+
"learning_rate": 4.7778839529767335e-05,
|
| 150 |
+
"loss": 0.5485,
|
| 151 |
+
"step": 2000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.62,
|
| 155 |
+
"eval_loss": 0.6321355700492859,
|
| 156 |
+
"eval_runtime": 524.3064,
|
| 157 |
+
"eval_samples_per_second": 12.056,
|
| 158 |
+
"eval_steps_per_second": 1.509,
|
| 159 |
+
"step": 2000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.68,
|
| 163 |
+
"grad_norm": 1.1281764507293701,
|
| 164 |
+
"learning_rate": 4.755661236916376e-05,
|
| 165 |
+
"loss": 0.5412,
|
| 166 |
+
"step": 2200
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.68,
|
| 170 |
+
"eval_loss": 0.6182820796966553,
|
| 171 |
+
"eval_runtime": 524.0853,
|
| 172 |
+
"eval_samples_per_second": 12.061,
|
| 173 |
+
"eval_steps_per_second": 1.509,
|
| 174 |
+
"step": 2200
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.74,
|
| 178 |
+
"grad_norm": 1.2166435718536377,
|
| 179 |
+
"learning_rate": 4.73343852085602e-05,
|
| 180 |
+
"loss": 0.5244,
|
| 181 |
+
"step": 2400
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.74,
|
| 185 |
+
"eval_loss": 0.6053413152694702,
|
| 186 |
+
"eval_runtime": 524.1727,
|
| 187 |
+
"eval_samples_per_second": 12.059,
|
| 188 |
+
"eval_steps_per_second": 1.509,
|
| 189 |
+
"step": 2400
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.8,
|
| 193 |
+
"grad_norm": 1.0197983980178833,
|
| 194 |
+
"learning_rate": 4.7112158047956625e-05,
|
| 195 |
+
"loss": 0.5161,
|
| 196 |
+
"step": 2600
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.8,
|
| 200 |
+
"eval_loss": 0.5936387181282043,
|
| 201 |
+
"eval_runtime": 524.5444,
|
| 202 |
+
"eval_samples_per_second": 12.05,
|
| 203 |
+
"eval_steps_per_second": 1.508,
|
| 204 |
+
"step": 2600
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.87,
|
| 208 |
+
"grad_norm": 1.0752451419830322,
|
| 209 |
+
"learning_rate": 4.688993088735306e-05,
|
| 210 |
+
"loss": 0.5043,
|
| 211 |
+
"step": 2800
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.87,
|
| 215 |
+
"eval_loss": 0.583480715751648,
|
| 216 |
+
"eval_runtime": 524.1844,
|
| 217 |
+
"eval_samples_per_second": 12.059,
|
| 218 |
+
"eval_steps_per_second": 1.509,
|
| 219 |
+
"step": 2800
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.93,
|
| 223 |
+
"grad_norm": 1.186328649520874,
|
| 224 |
+
"learning_rate": 4.666770372674949e-05,
|
| 225 |
+
"loss": 0.495,
|
| 226 |
+
"step": 3000
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.93,
|
| 230 |
+
"eval_loss": 0.5726756453514099,
|
| 231 |
+
"eval_runtime": 524.0268,
|
| 232 |
+
"eval_samples_per_second": 12.062,
|
| 233 |
+
"eval_steps_per_second": 1.509,
|
| 234 |
+
"step": 3000
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.99,
|
| 238 |
+
"grad_norm": 1.3003454208374023,
|
| 239 |
+
"learning_rate": 4.644547656614592e-05,
|
| 240 |
+
"loss": 0.484,
|
| 241 |
+
"step": 3200
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.99,
|
| 245 |
+
"eval_loss": 0.5635731816291809,
|
| 246 |
+
"eval_runtime": 523.978,
|
| 247 |
+
"eval_samples_per_second": 12.063,
|
| 248 |
+
"eval_steps_per_second": 1.51,
|
| 249 |
+
"step": 3200
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.05,
|
| 253 |
+
"grad_norm": 1.1711817979812622,
|
| 254 |
+
"learning_rate": 4.622324940554235e-05,
|
| 255 |
+
"loss": 0.4377,
|
| 256 |
+
"step": 3400
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.05,
|
| 260 |
+
"eval_loss": 0.5571883916854858,
|
| 261 |
+
"eval_runtime": 524.2962,
|
| 262 |
+
"eval_samples_per_second": 12.056,
|
| 263 |
+
"eval_steps_per_second": 1.509,
|
| 264 |
+
"step": 3400
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.11,
|
| 268 |
+
"grad_norm": 1.191895842552185,
|
| 269 |
+
"learning_rate": 4.6001022244938776e-05,
|
| 270 |
+
"loss": 0.427,
|
| 271 |
+
"step": 3600
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.11,
|
| 275 |
+
"eval_loss": 0.5517956614494324,
|
| 276 |
+
"eval_runtime": 524.0245,
|
| 277 |
+
"eval_samples_per_second": 12.062,
|
| 278 |
+
"eval_steps_per_second": 1.509,
|
| 279 |
+
"step": 3600
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.18,
|
| 283 |
+
"grad_norm": 1.1573776006698608,
|
| 284 |
+
"learning_rate": 4.577879508433521e-05,
|
| 285 |
+
"loss": 0.4216,
|
| 286 |
+
"step": 3800
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.18,
|
| 290 |
+
"eval_loss": 0.5458714962005615,
|
| 291 |
+
"eval_runtime": 523.9381,
|
| 292 |
+
"eval_samples_per_second": 12.064,
|
| 293 |
+
"eval_steps_per_second": 1.51,
|
| 294 |
+
"step": 3800
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.24,
|
| 298 |
+
"grad_norm": 1.232151985168457,
|
| 299 |
+
"learning_rate": 4.555656792373164e-05,
|
| 300 |
+
"loss": 0.4272,
|
| 301 |
+
"step": 4000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.24,
|
| 305 |
+
"eval_loss": 0.5407230257987976,
|
| 306 |
+
"eval_runtime": 523.9289,
|
| 307 |
+
"eval_samples_per_second": 12.065,
|
| 308 |
+
"eval_steps_per_second": 1.51,
|
| 309 |
+
"step": 4000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.3,
|
| 313 |
+
"grad_norm": 1.2596906423568726,
|
| 314 |
+
"learning_rate": 4.5334340763128066e-05,
|
| 315 |
+
"loss": 0.4197,
|
| 316 |
+
"step": 4200
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.3,
|
| 320 |
+
"eval_loss": 0.5343000888824463,
|
| 321 |
+
"eval_runtime": 524.1586,
|
| 322 |
+
"eval_samples_per_second": 12.059,
|
| 323 |
+
"eval_steps_per_second": 1.509,
|
| 324 |
+
"step": 4200
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.36,
|
| 328 |
+
"grad_norm": 1.1676970720291138,
|
| 329 |
+
"learning_rate": 4.51121136025245e-05,
|
| 330 |
+
"loss": 0.4188,
|
| 331 |
+
"step": 4400
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.36,
|
| 335 |
+
"eval_loss": 0.5275489687919617,
|
| 336 |
+
"eval_runtime": 523.8327,
|
| 337 |
+
"eval_samples_per_second": 12.067,
|
| 338 |
+
"eval_steps_per_second": 1.51,
|
| 339 |
+
"step": 4400
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.42,
|
| 343 |
+
"grad_norm": 1.291373372077942,
|
| 344 |
+
"learning_rate": 4.4889886441920935e-05,
|
| 345 |
+
"loss": 0.4165,
|
| 346 |
+
"step": 4600
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.42,
|
| 350 |
+
"eval_loss": 0.5239323377609253,
|
| 351 |
+
"eval_runtime": 523.8294,
|
| 352 |
+
"eval_samples_per_second": 12.067,
|
| 353 |
+
"eval_steps_per_second": 1.51,
|
| 354 |
+
"step": 4600
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.48,
|
| 358 |
+
"grad_norm": 1.3542147874832153,
|
| 359 |
+
"learning_rate": 4.466765928131736e-05,
|
| 360 |
+
"loss": 0.411,
|
| 361 |
+
"step": 4800
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.48,
|
| 365 |
+
"eval_loss": 0.5183799862861633,
|
| 366 |
+
"eval_runtime": 523.956,
|
| 367 |
+
"eval_samples_per_second": 12.064,
|
| 368 |
+
"eval_steps_per_second": 1.51,
|
| 369 |
+
"step": 4800
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.55,
|
| 373 |
+
"grad_norm": 1.138175368309021,
|
| 374 |
+
"learning_rate": 4.4445432120713797e-05,
|
| 375 |
+
"loss": 0.4066,
|
| 376 |
+
"step": 5000
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 1.55,
|
| 380 |
+
"eval_loss": 0.5129222869873047,
|
| 381 |
+
"eval_runtime": 524.4918,
|
| 382 |
+
"eval_samples_per_second": 12.052,
|
| 383 |
+
"eval_steps_per_second": 1.508,
|
| 384 |
+
"step": 5000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.61,
|
| 388 |
+
"grad_norm": 1.2202116250991821,
|
| 389 |
+
"learning_rate": 4.4223204960110224e-05,
|
| 390 |
+
"loss": 0.4054,
|
| 391 |
+
"step": 5200
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.61,
|
| 395 |
+
"eval_loss": 0.5082316398620605,
|
| 396 |
+
"eval_runtime": 524.0745,
|
| 397 |
+
"eval_samples_per_second": 12.061,
|
| 398 |
+
"eval_steps_per_second": 1.509,
|
| 399 |
+
"step": 5200
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.67,
|
| 403 |
+
"grad_norm": 1.266581654548645,
|
| 404 |
+
"learning_rate": 4.400097779950666e-05,
|
| 405 |
+
"loss": 0.4015,
|
| 406 |
+
"step": 5400
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.67,
|
| 410 |
+
"eval_loss": 0.5040789842605591,
|
| 411 |
+
"eval_runtime": 524.5202,
|
| 412 |
+
"eval_samples_per_second": 12.051,
|
| 413 |
+
"eval_steps_per_second": 1.508,
|
| 414 |
+
"step": 5400
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.73,
|
| 418 |
+
"grad_norm": 1.27601957321167,
|
| 419 |
+
"learning_rate": 4.3778750638903086e-05,
|
| 420 |
+
"loss": 0.3982,
|
| 421 |
+
"step": 5600
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.73,
|
| 425 |
+
"eval_loss": 0.5011813044548035,
|
| 426 |
+
"eval_runtime": 524.275,
|
| 427 |
+
"eval_samples_per_second": 12.057,
|
| 428 |
+
"eval_steps_per_second": 1.509,
|
| 429 |
+
"step": 5600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.79,
|
| 433 |
+
"grad_norm": 1.2219188213348389,
|
| 434 |
+
"learning_rate": 4.355652347829952e-05,
|
| 435 |
+
"loss": 0.3954,
|
| 436 |
+
"step": 5800
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.79,
|
| 440 |
+
"eval_loss": 0.49553418159484863,
|
| 441 |
+
"eval_runtime": 524.2405,
|
| 442 |
+
"eval_samples_per_second": 12.057,
|
| 443 |
+
"eval_steps_per_second": 1.509,
|
| 444 |
+
"step": 5800
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.86,
|
| 448 |
+
"grad_norm": 1.0724387168884277,
|
| 449 |
+
"learning_rate": 4.333429631769595e-05,
|
| 450 |
+
"loss": 0.4003,
|
| 451 |
+
"step": 6000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.86,
|
| 455 |
+
"eval_loss": 0.4928473234176636,
|
| 456 |
+
"eval_runtime": 524.0586,
|
| 457 |
+
"eval_samples_per_second": 12.062,
|
| 458 |
+
"eval_steps_per_second": 1.509,
|
| 459 |
+
"step": 6000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.92,
|
| 463 |
+
"grad_norm": 1.1879562139511108,
|
| 464 |
+
"learning_rate": 4.311206915709238e-05,
|
| 465 |
+
"loss": 0.3925,
|
| 466 |
+
"step": 6200
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 1.92,
|
| 470 |
+
"eval_loss": 0.4879492521286011,
|
| 471 |
+
"eval_runtime": 523.9744,
|
| 472 |
+
"eval_samples_per_second": 12.064,
|
| 473 |
+
"eval_steps_per_second": 1.51,
|
| 474 |
+
"step": 6200
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.98,
|
| 478 |
+
"grad_norm": 1.207940697669983,
|
| 479 |
+
"learning_rate": 4.288984199648881e-05,
|
| 480 |
+
"loss": 0.3934,
|
| 481 |
+
"step": 6400
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 1.98,
|
| 485 |
+
"eval_loss": 0.48524412512779236,
|
| 486 |
+
"eval_runtime": 523.9417,
|
| 487 |
+
"eval_samples_per_second": 12.064,
|
| 488 |
+
"eval_steps_per_second": 1.51,
|
| 489 |
+
"step": 6400
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.04,
|
| 493 |
+
"grad_norm": 1.1332272291183472,
|
| 494 |
+
"learning_rate": 4.2667614835885244e-05,
|
| 495 |
+
"loss": 0.3432,
|
| 496 |
+
"step": 6600
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 2.04,
|
| 500 |
+
"eval_loss": 0.4873977303504944,
|
| 501 |
+
"eval_runtime": 523.9543,
|
| 502 |
+
"eval_samples_per_second": 12.064,
|
| 503 |
+
"eval_steps_per_second": 1.51,
|
| 504 |
+
"step": 6600
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.1,
|
| 508 |
+
"grad_norm": 1.1995199918746948,
|
| 509 |
+
"learning_rate": 4.244538767528167e-05,
|
| 510 |
+
"loss": 0.3242,
|
| 511 |
+
"step": 6800
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 2.1,
|
| 515 |
+
"eval_loss": 0.48586371541023254,
|
| 516 |
+
"eval_runtime": 524.0976,
|
| 517 |
+
"eval_samples_per_second": 12.061,
|
| 518 |
+
"eval_steps_per_second": 1.509,
|
| 519 |
+
"step": 6800
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 2.17,
|
| 523 |
+
"grad_norm": 1.3645168542861938,
|
| 524 |
+
"learning_rate": 4.2223160514678106e-05,
|
| 525 |
+
"loss": 0.3249,
|
| 526 |
+
"step": 7000
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 2.17,
|
| 530 |
+
"eval_loss": 0.4856572449207306,
|
| 531 |
+
"eval_runtime": 525.0728,
|
| 532 |
+
"eval_samples_per_second": 12.038,
|
| 533 |
+
"eval_steps_per_second": 1.506,
|
| 534 |
+
"step": 7000
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.23,
|
| 538 |
+
"grad_norm": 1.2310686111450195,
|
| 539 |
+
"learning_rate": 4.200093335407454e-05,
|
| 540 |
+
"loss": 0.3269,
|
| 541 |
+
"step": 7200
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 2.23,
|
| 545 |
+
"eval_loss": 0.4839836359024048,
|
| 546 |
+
"eval_runtime": 524.2734,
|
| 547 |
+
"eval_samples_per_second": 12.057,
|
| 548 |
+
"eval_steps_per_second": 1.509,
|
| 549 |
+
"step": 7200
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 2.29,
|
| 553 |
+
"grad_norm": 1.1063857078552246,
|
| 554 |
+
"learning_rate": 4.177870619347097e-05,
|
| 555 |
+
"loss": 0.3208,
|
| 556 |
+
"step": 7400
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 2.29,
|
| 560 |
+
"eval_loss": 0.4794396460056305,
|
| 561 |
+
"eval_runtime": 524.0289,
|
| 562 |
+
"eval_samples_per_second": 12.062,
|
| 563 |
+
"eval_steps_per_second": 1.509,
|
| 564 |
+
"step": 7400
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.35,
|
| 568 |
+
"grad_norm": 1.259097695350647,
|
| 569 |
+
"learning_rate": 4.15564790328674e-05,
|
| 570 |
+
"loss": 0.3285,
|
| 571 |
+
"step": 7600
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.35,
|
| 575 |
+
"eval_loss": 0.4764421582221985,
|
| 576 |
+
"eval_runtime": 523.9713,
|
| 577 |
+
"eval_samples_per_second": 12.064,
|
| 578 |
+
"eval_steps_per_second": 1.51,
|
| 579 |
+
"step": 7600
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 2.41,
|
| 583 |
+
"grad_norm": 1.4079738855361938,
|
| 584 |
+
"learning_rate": 4.133425187226383e-05,
|
| 585 |
+
"loss": 0.3298,
|
| 586 |
+
"step": 7800
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 2.41,
|
| 590 |
+
"eval_loss": 0.4742029011249542,
|
| 591 |
+
"eval_runtime": 525.3854,
|
| 592 |
+
"eval_samples_per_second": 12.031,
|
| 593 |
+
"eval_steps_per_second": 1.506,
|
| 594 |
+
"step": 7800
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 2.47,
|
| 598 |
+
"grad_norm": 1.3066357374191284,
|
| 599 |
+
"learning_rate": 4.1112024711660265e-05,
|
| 600 |
+
"loss": 0.3274,
|
| 601 |
+
"step": 8000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 2.47,
|
| 605 |
+
"eval_loss": 0.4709968864917755,
|
| 606 |
+
"eval_runtime": 524.0046,
|
| 607 |
+
"eval_samples_per_second": 12.063,
|
| 608 |
+
"eval_steps_per_second": 1.51,
|
| 609 |
+
"step": 8000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 2.54,
|
| 613 |
+
"grad_norm": 1.236858606338501,
|
| 614 |
+
"learning_rate": 4.088979755105669e-05,
|
| 615 |
+
"loss": 0.3279,
|
| 616 |
+
"step": 8200
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 2.54,
|
| 620 |
+
"eval_loss": 0.4686913788318634,
|
| 621 |
+
"eval_runtime": 524.0329,
|
| 622 |
+
"eval_samples_per_second": 12.062,
|
| 623 |
+
"eval_steps_per_second": 1.509,
|
| 624 |
+
"step": 8200
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 2.6,
|
| 628 |
+
"grad_norm": 1.256058931350708,
|
| 629 |
+
"learning_rate": 4.066757039045313e-05,
|
| 630 |
+
"loss": 0.3278,
|
| 631 |
+
"step": 8400
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 2.6,
|
| 635 |
+
"eval_loss": 0.46550899744033813,
|
| 636 |
+
"eval_runtime": 523.9744,
|
| 637 |
+
"eval_samples_per_second": 12.064,
|
| 638 |
+
"eval_steps_per_second": 1.51,
|
| 639 |
+
"step": 8400
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.66,
|
| 643 |
+
"grad_norm": 1.3171123266220093,
|
| 644 |
+
"learning_rate": 4.0445343229849554e-05,
|
| 645 |
+
"loss": 0.325,
|
| 646 |
+
"step": 8600
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.66,
|
| 650 |
+
"eval_loss": 0.46436843276023865,
|
| 651 |
+
"eval_runtime": 524.213,
|
| 652 |
+
"eval_samples_per_second": 12.058,
|
| 653 |
+
"eval_steps_per_second": 1.509,
|
| 654 |
+
"step": 8600
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 2.72,
|
| 658 |
+
"grad_norm": 1.095387578010559,
|
| 659 |
+
"learning_rate": 4.022311606924599e-05,
|
| 660 |
+
"loss": 0.329,
|
| 661 |
+
"step": 8800
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 2.72,
|
| 665 |
+
"eval_loss": 0.4598775804042816,
|
| 666 |
+
"eval_runtime": 524.0933,
|
| 667 |
+
"eval_samples_per_second": 12.061,
|
| 668 |
+
"eval_steps_per_second": 1.509,
|
| 669 |
+
"step": 8800
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.78,
|
| 673 |
+
"grad_norm": 1.1329820156097412,
|
| 674 |
+
"learning_rate": 4.0000888908642416e-05,
|
| 675 |
+
"loss": 0.3281,
|
| 676 |
+
"step": 9000
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.78,
|
| 680 |
+
"eval_loss": 0.4569958746433258,
|
| 681 |
+
"eval_runtime": 524.501,
|
| 682 |
+
"eval_samples_per_second": 12.051,
|
| 683 |
+
"eval_steps_per_second": 1.508,
|
| 684 |
+
"step": 9000
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 2.85,
|
| 688 |
+
"grad_norm": 1.1984633207321167,
|
| 689 |
+
"learning_rate": 3.977866174803885e-05,
|
| 690 |
+
"loss": 0.3301,
|
| 691 |
+
"step": 9200
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 2.85,
|
| 695 |
+
"eval_loss": 0.4568074643611908,
|
| 696 |
+
"eval_runtime": 524.1689,
|
| 697 |
+
"eval_samples_per_second": 12.059,
|
| 698 |
+
"eval_steps_per_second": 1.509,
|
| 699 |
+
"step": 9200
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 2.91,
|
| 703 |
+
"grad_norm": 1.2546519041061401,
|
| 704 |
+
"learning_rate": 3.955643458743528e-05,
|
| 705 |
+
"loss": 0.3329,
|
| 706 |
+
"step": 9400
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 2.91,
|
| 710 |
+
"eval_loss": 0.4547139108181,
|
| 711 |
+
"eval_runtime": 523.8736,
|
| 712 |
+
"eval_samples_per_second": 12.066,
|
| 713 |
+
"eval_steps_per_second": 1.51,
|
| 714 |
+
"step": 9400
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 2.97,
|
| 718 |
+
"grad_norm": 1.077413558959961,
|
| 719 |
+
"learning_rate": 3.933420742683171e-05,
|
| 720 |
+
"loss": 0.3252,
|
| 721 |
+
"step": 9600
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 2.97,
|
| 725 |
+
"eval_loss": 0.45176568627357483,
|
| 726 |
+
"eval_runtime": 523.866,
|
| 727 |
+
"eval_samples_per_second": 12.066,
|
| 728 |
+
"eval_steps_per_second": 1.51,
|
| 729 |
+
"step": 9600
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 3.03,
|
| 733 |
+
"grad_norm": 1.446174144744873,
|
| 734 |
+
"learning_rate": 3.911198026622814e-05,
|
| 735 |
+
"loss": 0.2923,
|
| 736 |
+
"step": 9800
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 3.03,
|
| 740 |
+
"eval_loss": 0.46238696575164795,
|
| 741 |
+
"eval_runtime": 524.0417,
|
| 742 |
+
"eval_samples_per_second": 12.062,
|
| 743 |
+
"eval_steps_per_second": 1.509,
|
| 744 |
+
"step": 9800
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 3.09,
|
| 748 |
+
"grad_norm": 1.3876376152038574,
|
| 749 |
+
"learning_rate": 3.8889753105624575e-05,
|
| 750 |
+
"loss": 0.2567,
|
| 751 |
+
"step": 10000
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 3.09,
|
| 755 |
+
"eval_loss": 0.46406450867652893,
|
| 756 |
+
"eval_runtime": 523.8474,
|
| 757 |
+
"eval_samples_per_second": 12.066,
|
| 758 |
+
"eval_steps_per_second": 1.51,
|
| 759 |
+
"step": 10000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 3.15,
|
| 763 |
+
"grad_norm": 1.3354049921035767,
|
| 764 |
+
"learning_rate": 3.8667525945021e-05,
|
| 765 |
+
"loss": 0.2639,
|
| 766 |
+
"step": 10200
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 3.15,
|
| 770 |
+
"eval_loss": 0.46170929074287415,
|
| 771 |
+
"eval_runtime": 523.6743,
|
| 772 |
+
"eval_samples_per_second": 12.07,
|
| 773 |
+
"eval_steps_per_second": 1.51,
|
| 774 |
+
"step": 10200
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 3.22,
|
| 778 |
+
"grad_norm": 1.3357325792312622,
|
| 779 |
+
"learning_rate": 3.844529878441743e-05,
|
| 780 |
+
"loss": 0.2679,
|
| 781 |
+
"step": 10400
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 3.22,
|
| 785 |
+
"eval_loss": 0.462502658367157,
|
| 786 |
+
"eval_runtime": 523.7861,
|
| 787 |
+
"eval_samples_per_second": 12.068,
|
| 788 |
+
"eval_steps_per_second": 1.51,
|
| 789 |
+
"step": 10400
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 3.28,
|
| 793 |
+
"grad_norm": 1.2143915891647339,
|
| 794 |
+
"learning_rate": 3.8223071623813864e-05,
|
| 795 |
+
"loss": 0.2678,
|
| 796 |
+
"step": 10600
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 3.28,
|
| 800 |
+
"eval_loss": 0.461648166179657,
|
| 801 |
+
"eval_runtime": 523.8835,
|
| 802 |
+
"eval_samples_per_second": 12.066,
|
| 803 |
+
"eval_steps_per_second": 1.51,
|
| 804 |
+
"step": 10600
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 3.34,
|
| 808 |
+
"grad_norm": 1.4215404987335205,
|
| 809 |
+
"learning_rate": 3.800084446321029e-05,
|
| 810 |
+
"loss": 0.272,
|
| 811 |
+
"step": 10800
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 3.34,
|
| 815 |
+
"eval_loss": 0.45769479870796204,
|
| 816 |
+
"eval_runtime": 524.0212,
|
| 817 |
+
"eval_samples_per_second": 12.062,
|
| 818 |
+
"eval_steps_per_second": 1.509,
|
| 819 |
+
"step": 10800
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 3.4,
|
| 823 |
+
"grad_norm": 1.258699893951416,
|
| 824 |
+
"learning_rate": 3.7778617302606726e-05,
|
| 825 |
+
"loss": 0.2734,
|
| 826 |
+
"step": 11000
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 3.4,
|
| 830 |
+
"eval_loss": 0.45607540011405945,
|
| 831 |
+
"eval_runtime": 523.6473,
|
| 832 |
+
"eval_samples_per_second": 12.071,
|
| 833 |
+
"eval_steps_per_second": 1.511,
|
| 834 |
+
"step": 11000
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 3.46,
|
| 838 |
+
"grad_norm": 1.329528570175171,
|
| 839 |
+
"learning_rate": 3.7556390142003154e-05,
|
| 840 |
+
"loss": 0.2744,
|
| 841 |
+
"step": 11200
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 3.46,
|
| 845 |
+
"eval_loss": 0.4543399214744568,
|
| 846 |
+
"eval_runtime": 523.8153,
|
| 847 |
+
"eval_samples_per_second": 12.067,
|
| 848 |
+
"eval_steps_per_second": 1.51,
|
| 849 |
+
"step": 11200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 3.53,
|
| 853 |
+
"grad_norm": 1.265796184539795,
|
| 854 |
+
"learning_rate": 3.733416298139959e-05,
|
| 855 |
+
"loss": 0.2727,
|
| 856 |
+
"step": 11400
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 3.53,
|
| 860 |
+
"eval_loss": 0.4526574909687042,
|
| 861 |
+
"eval_runtime": 523.7836,
|
| 862 |
+
"eval_samples_per_second": 12.068,
|
| 863 |
+
"eval_steps_per_second": 1.51,
|
| 864 |
+
"step": 11400
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 3.59,
|
| 868 |
+
"grad_norm": 1.3111358880996704,
|
| 869 |
+
"learning_rate": 3.7111935820796016e-05,
|
| 870 |
+
"loss": 0.2723,
|
| 871 |
+
"step": 11600
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 3.59,
|
| 875 |
+
"eval_loss": 0.452004998922348,
|
| 876 |
+
"eval_runtime": 523.8302,
|
| 877 |
+
"eval_samples_per_second": 12.067,
|
| 878 |
+
"eval_steps_per_second": 1.51,
|
| 879 |
+
"step": 11600
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 3.65,
|
| 883 |
+
"grad_norm": 1.2918260097503662,
|
| 884 |
+
"learning_rate": 3.688970866019245e-05,
|
| 885 |
+
"loss": 0.2727,
|
| 886 |
+
"step": 11800
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 3.65,
|
| 890 |
+
"eval_loss": 0.4493946433067322,
|
| 891 |
+
"eval_runtime": 523.8682,
|
| 892 |
+
"eval_samples_per_second": 12.066,
|
| 893 |
+
"eval_steps_per_second": 1.51,
|
| 894 |
+
"step": 11800
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 3.71,
|
| 898 |
+
"grad_norm": 1.3306081295013428,
|
| 899 |
+
"learning_rate": 3.666748149958888e-05,
|
| 900 |
+
"loss": 0.2759,
|
| 901 |
+
"step": 12000
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 3.71,
|
| 905 |
+
"eval_loss": 0.4486713707447052,
|
| 906 |
+
"eval_runtime": 524.0531,
|
| 907 |
+
"eval_samples_per_second": 12.062,
|
| 908 |
+
"eval_steps_per_second": 1.509,
|
| 909 |
+
"step": 12000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 3.77,
|
| 913 |
+
"grad_norm": 1.253534197807312,
|
| 914 |
+
"learning_rate": 3.644525433898531e-05,
|
| 915 |
+
"loss": 0.2727,
|
| 916 |
+
"step": 12200
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 3.77,
|
| 920 |
+
"eval_loss": 0.44542327523231506,
|
| 921 |
+
"eval_runtime": 524.1367,
|
| 922 |
+
"eval_samples_per_second": 12.06,
|
| 923 |
+
"eval_steps_per_second": 1.509,
|
| 924 |
+
"step": 12200
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 3.84,
|
| 928 |
+
"grad_norm": 1.1979799270629883,
|
| 929 |
+
"learning_rate": 3.622302717838174e-05,
|
| 930 |
+
"loss": 0.2747,
|
| 931 |
+
"step": 12400
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 3.84,
|
| 935 |
+
"eval_loss": 0.44225621223449707,
|
| 936 |
+
"eval_runtime": 523.5918,
|
| 937 |
+
"eval_samples_per_second": 12.072,
|
| 938 |
+
"eval_steps_per_second": 1.511,
|
| 939 |
+
"step": 12400
|
| 940 |
+
}
|
| 941 |
+
],
|
| 942 |
+
"logging_steps": 200,
|
| 943 |
+
"max_steps": 45000,
|
| 944 |
+
"num_input_tokens_seen": 0,
|
| 945 |
+
"num_train_epochs": 14,
|
| 946 |
+
"save_steps": 400,
|
| 947 |
+
"total_flos": 8.769484235678417e+18,
|
| 948 |
+
"train_batch_size": 64,
|
| 949 |
+
"trial_name": null,
|
| 950 |
+
"trial_params": null
|
| 951 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d50493b56d5533a9e1b2cd97c4c0cb4476281e6bddf9c1c5ce2fd2538909c88
|
| 3 |
+
size 4920
|