End of training

README.md

@@ -6,7 +6,7 @@ tags:
 - axolotl
 - generated_from_trainer
 model-index:
-- name: 118ed5ed-7122-4885-96b2-48d099cf26b6
+- name: 68938adc-cdd4-4d29-81d8-8c7ffc6c6fc3
   results: []
 ---
 
@@ -18,16 +18,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 axolotl version: `0.4.1`
 ```yaml
-accelerate_config:
-  dynamo_backend: inductor
-  mixed_precision: bf16
-  num_machines: 1
-  num_processes: auto
-  use_cpu: false
 adapter: lora
 base_model: unsloth/mistral-7b
 bf16: auto
 chat_template: llama3
+cosine_min_lr_ratio: 0.1
+data_processes: 16
 dataset_prepared_path: null
 datasets:
 - data_files:
@@ -45,75 +41,88 @@ datasets:
     system_prompt: ''
 debug: null
 deepspeed: null
-device_map: auto
-early_stopping_patience: null
-eval_max_new_tokens: 128
-eval_table_size: null
-evals_per_epoch: 4
+device_map: '{'''':torch.cuda.current_device()}'
+do_eval: true
+early_stopping_patience: 1
+eval_batch_size: 1
+eval_sample_packing: false
+eval_steps: 25
+evaluation_strategy: steps
 flash_attention: false
 fp16: null
 fsdp: null
 fsdp_config: null
-gradient_accumulation_steps: 16
+gradient_accumulation_steps: 32
 gradient_checkpointing: true
-group_by_length: false
+group_by_length: true
 hub_model_id: null
-hub_repo: null
+hub_repo: stevemonite
 hub_strategy: checkpoint
 hub_token: null
-learning_rate: 0.0001
+learning_rate: 0.0003
+load_in_4bit: false
+load_in_8bit: false
 local_rank: null
 logging_steps: 1
-lora_alpha: 16
+lora_alpha: 64
 lora_dropout: 0.05
 lora_fan_in_fan_out: null
 lora_model_dir: null
-lora_r: 8
+lora_r: 32
 lora_target_linear: true
 lora_target_modules:
 - q_proj
+- k_proj
 - v_proj
+- o_proj
+- gate_proj
+- down_proj
+- up_proj
 lr_scheduler: cosine
+max_grad_norm: 1.0
 max_memory:
   0: 70GiB
-max_steps: 100
-micro_batch_size: 2
+max_steps: 76
+micro_batch_size: 1
 mlflow_experiment_name: /tmp/8b2f2e480f49b204_train_data.json
 model_type: AutoModelForCausalLM
-num_epochs: 1
-optimizer: adamw_bnb_8bit
+num_epochs: 2
+optim_args:
+  adam_beta1: 0.9
+  adam_beta2: 0.95
+  adam_epsilon: 1e-5
+optimizer: adamw_torch
 output_dir: miner_id_24
 pad_to_sequence_len: true
-quantization_config:
-  llm_int8_enable_fp32_cpu_offload: true
-  load_in_8bit: true
 resume_from_checkpoint: null
 s2_attention: null
 sample_packing: false
-saves_per_epoch: 4
-sequence_len: 512
+save_steps: 50
+save_strategy: steps
+sequence_len: 2048
 strict: false
 tf32: false
 tokenizer_type: AutoTokenizer
-torch_compile: true
+torch_compile: false
 train_on_inputs: false
 trust_remote_code: true
-val_set_size: 0.05
-wandb_entity: null
-wandb_mode: online
-wandb_name: 118ed5ed-7122-4885-96b2-48d099cf26b6
-wandb_project: Gradients-On-Demand
-wandb_run: your_name
-wandb_runid: 118ed5ed-7122-4885-96b2-48d099cf26b6
-warmup_steps: 10
-weight_decay: 0.0
+val_set_size: 50
+wandb_entity: sn56-miner
+wandb_mode: disabled
+wandb_name: sn56d1/6e0123f2
+wandb_project: god
+wandb_run: pvxz
+wandb_runid: sn56d1/6e0123f2
+warmup_raio: 0.03
+warmup_ratio: 0.03
+weight_decay: 0.01
 xformers_attention: null
 
 ```
 
 </details><br>
 
-# 118ed5ed-7122-4885-96b2-48d099cf26b6
+# 68938adc-cdd4-4d29-81d8-8c7ffc6c6fc3
 
 This model is a fine-tuned version of [unsloth/mistral-7b](https://huggingface.co/unsloth/mistral-7b) on the None dataset.
 It achieves the following results on the evaluation set:
@@ -136,26 +145,28 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0001
-- train_batch_size: 2
-- eval_batch_size: 2
+- learning_rate: 0.0003
+- train_batch_size: 1
+- eval_batch_size: 1
 - seed: 42
-- gradient_accumulation_steps: 16
-- total_train_batch_size: 32
-- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 32
+- total_train_batch_size: 128
+- total_eval_batch_size: 4
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5
 - lr_scheduler_type: cosine
-- lr_scheduler_warmup_steps: 10
-- training_steps: 100
+- lr_scheduler_warmup_steps: 2
+- training_steps: 76
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 0.0           | 0.0007 | 1    | nan             |
-| 0.0           | 0.0169 | 25   | nan             |
-| 0.0           | 0.0338 | 50   | nan             |
-| 0.0           | 0.0507 | 75   | nan             |
-| 0.0           | 0.0676 | 100  | nan             |
+| 0.0           | 0.0026 | 1    | nan             |
+| 0.0           | 0.0643 | 25   | nan             |
+| 0.0           | 0.1286 | 50   | nan             |
+| 0.0           | 0.1930 | 75   | nan             |
 
 
 ### Framework versions

adapter_config.json

@@ -10,23 +10,23 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 16,
+  "lora_alpha": 64,
   "lora_dropout": 0.05,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 8,
+  "r": 32,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "o_proj",
-    "v_proj",
-    "down_proj",
+    "gate_proj",
     "k_proj",
+    "o_proj",
     "up_proj",
-    "gate_proj"
+    "q_proj",
+    "v_proj",
+    "down_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:585178d5250a9068d12faf659c980689a027ffee93b57f3d772b132f04394541
-size 84047370
+oid sha256:0e1e22a7c24473ae2ce03415b098f515aeed520202a76c93116fabb734bf0b47
+size 335706186

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec79eaa4b64976106e2d19432a45f0ab1bec56041129a34042d5cd263efa55ec
-size 83945296
+oid sha256:6c696cab0a3d3a6350c331a2fa51b164096d09045d83c85832605ebb6c37a08a
+size 335604696

last-checkpoint/adapter_config.json

@@ -10,23 +10,23 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 16,
+  "lora_alpha": 64,
   "lora_dropout": 0.05,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 8,
+  "r": 32,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "o_proj",
-    "v_proj",
-    "down_proj",
+    "gate_proj",
     "k_proj",
+    "o_proj",
     "up_proj",
-    "gate_proj"
+    "q_proj",
+    "v_proj",
+    "down_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec79eaa4b64976106e2d19432a45f0ab1bec56041129a34042d5cd263efa55ec
-size 83945296
+oid sha256:6c696cab0a3d3a6350c331a2fa51b164096d09045d83c85832605ebb6c37a08a
+size 335604696

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:06ee945e16ae22c9075277531a6f8bc0ec708f7e7d0f7f18b4a9e5b5ec85eb27
-size 43122580
+oid sha256:4045837df223652ae1807aca3bdda171754240535f50bc4560efaea4d8ea8d2b
+size 671466706

last-checkpoint/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33b8177a44a181e058c6946682eb2e19c1ff617f6f261ed3c51a5c6f4afe7589
+size 14960

last-checkpoint/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92b88ed4526213f4665a86c47810ec278c7eb6676845fd971c653281bb156b69
+size 15024

last-checkpoint/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed74ec87b52d4f93bcb2e178d6d91b0136149eac291bb360c8bc1d5d9f955ebf
+size 15024

last-checkpoint/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9cd698697de34401eb1d1493fed0acbea22d6de9691b03adf380d3e3dda735ab
+size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:49d60a69e2379be2053e816cbaff31e6c931b5922dd86c71c9eaf473299cbf62
+oid sha256:eff4e51807fc2798ef7e3176a7ae7a505f7b610c949cf8c1d65662116e17ea14
 size 1064

last-checkpoint/trainer_state.json

@@ -1,773 +1,416 @@
 {
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 0.06763897696047347,
+  "best_metric": NaN,
+  "best_model_checkpoint": "miner_id_24/checkpoint-50",
+  "epoch": 0.12863804470172052,
   "eval_steps": 25,
-  "global_step": 100,
+  "global_step": 50,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.0006763897696047348,
+      "epoch": 0.0025727608940344106,
       "grad_norm": NaN,
-      "learning_rate": 1e-05,
+      "learning_rate": 0.00015,
       "loss": 0.0,
       "step": 1
     },
     {
-      "epoch": 0.0006763897696047348,
+      "epoch": 0.0025727608940344106,
       "eval_loss": NaN,
-      "eval_runtime": 211.3876,
-      "eval_samples_per_second": 11.779,
-      "eval_steps_per_second": 5.89,
+      "eval_runtime": 1.4827,
+      "eval_samples_per_second": 33.722,
+      "eval_steps_per_second": 8.768,
       "step": 1
     },
     {
-      "epoch": 0.0013527795392094695,
+      "epoch": 0.005145521788068821,
       "grad_norm": NaN,
-      "learning_rate": 2e-05,
+      "learning_rate": 0.0003,
       "loss": 0.0,
       "step": 2
     },
     {
-      "epoch": 0.002029169308814204,
+      "epoch": 0.007718282682103232,
       "grad_norm": NaN,
-      "learning_rate": 3e-05,
+      "learning_rate": 0.000299878360437632,
       "loss": 0.0,
       "step": 3
     },
     {
-      "epoch": 0.002705559078418939,
+      "epoch": 0.010291043576137643,
       "grad_norm": NaN,
-      "learning_rate": 4e-05,
+      "learning_rate": 0.00029951366095324104,
       "loss": 0.0,
       "step": 4
     },
     {
-      "epoch": 0.0033819488480236735,
+      "epoch": 0.012863804470172054,
       "grad_norm": NaN,
-      "learning_rate": 5e-05,
+      "learning_rate": 0.00029890655875994835,
       "loss": 0.0,
       "step": 5
     },
     {
-      "epoch": 0.004058338617628408,
+      "epoch": 0.015436565364206465,
       "grad_norm": NaN,
-      "learning_rate": 6e-05,
+      "learning_rate": 0.0002980581478969406,
       "loss": 0.0,
       "step": 6
     },
     {
-      "epoch": 0.004734728387233143,
+      "epoch": 0.018009326258240876,
       "grad_norm": NaN,
-      "learning_rate": 7e-05,
+      "learning_rate": 0.00029696995725793764,
       "loss": 0.0,
       "step": 7
     },
     {
-      "epoch": 0.005411118156837878,
+      "epoch": 0.020582087152275285,
       "grad_norm": NaN,
-      "learning_rate": 8e-05,
+      "learning_rate": 0.00029564394783602234,
       "loss": 0.0,
       "step": 8
     },
     {
-      "epoch": 0.0060875079264426125,
+      "epoch": 0.023154848046309694,
       "grad_norm": NaN,
-      "learning_rate": 9e-05,
+      "learning_rate": 0.0002940825091897988,
       "loss": 0.0,
       "step": 9
     },
     {
-      "epoch": 0.006763897696047347,
+      "epoch": 0.025727608940344107,
       "grad_norm": NaN,
-      "learning_rate": 0.0001,
+      "learning_rate": 0.00029228845513724634,
       "loss": 0.0,
       "step": 10
     },
     {
-      "epoch": 0.007440287465652082,
+      "epoch": 0.028300369834378516,
       "grad_norm": NaN,
-      "learning_rate": 9.99695413509548e-05,
+      "learning_rate": 0.00029026501868502873,
       "loss": 0.0,
       "step": 11
     },
     {
-      "epoch": 0.008116677235256817,
+      "epoch": 0.03087313072841293,
       "grad_norm": NaN,
-      "learning_rate": 9.987820251299122e-05,
+      "learning_rate": 0.0002880158462023983,
       "loss": 0.0,
       "step": 12
     },
     {
-      "epoch": 0.008793067004861552,
+      "epoch": 0.03344589162244734,
       "grad_norm": NaN,
-      "learning_rate": 9.972609476841367e-05,
+      "learning_rate": 0.0002855449908501917,
       "loss": 0.0,
       "step": 13
     },
     {
-      "epoch": 0.009469456774466286,
+      "epoch": 0.03601865251648175,
       "grad_norm": NaN,
-      "learning_rate": 9.951340343707852e-05,
+      "learning_rate": 0.00028285690527676035,
       "loss": 0.0,
       "step": 14
     },
     {
-      "epoch": 0.010145846544071021,
+      "epoch": 0.03859141341051616,
       "grad_norm": NaN,
-      "learning_rate": 9.924038765061042e-05,
+      "learning_rate": 0.000279956433593997,
       "loss": 0.0,
       "step": 15
     },
     {
-      "epoch": 0.010822236313675756,
+      "epoch": 0.04116417430455057,
       "grad_norm": NaN,
-      "learning_rate": 9.890738003669029e-05,
+      "learning_rate": 0.00027684880264791867,
       "loss": 0.0,
       "step": 16
     },
     {
-      "epoch": 0.01149862608328049,
+      "epoch": 0.04373693519858498,
       "grad_norm": NaN,
-      "learning_rate": 9.851478631379982e-05,
+      "learning_rate": 0.00027353961259953696,
       "loss": 0.0,
       "step": 17
     },
     {
-      "epoch": 0.012175015852885225,
+      "epoch": 0.04630969609261939,
       "grad_norm": NaN,
-      "learning_rate": 9.806308479691595e-05,
+      "learning_rate": 0.00027003482683298933,
       "loss": 0.0,
       "step": 18
     },
     {
-      "epoch": 0.01285140562248996,
+      "epoch": 0.048882456986653805,
       "grad_norm": NaN,
-      "learning_rate": 9.755282581475769e-05,
+      "learning_rate": 0.00026634076120911777,
       "loss": 0.0,
       "step": 19
     },
     {
-      "epoch": 0.013527795392094694,
+      "epoch": 0.051455217880688214,
       "grad_norm": NaN,
-      "learning_rate": 9.698463103929542e-05,
+      "learning_rate": 0.0002624640726838608,
       "loss": 0.0,
       "step": 20
     },
     {
-      "epoch": 0.01420418516169943,
+      "epoch": 0.054027978774722624,
       "grad_norm": NaN,
-      "learning_rate": 9.635919272833938e-05,
+      "learning_rate": 0.00025841174731196877,
       "loss": 0.0,
       "step": 21
     },
     {
-      "epoch": 0.014880574931304165,
+      "epoch": 0.05660073966875703,
       "grad_norm": NaN,
-      "learning_rate": 9.567727288213005e-05,
+      "learning_rate": 0.000254191087657661,
       "loss": 0.0,
       "step": 22
     },
     {
-      "epoch": 0.015556964700908898,
+      "epoch": 0.05917350056279144,
       "grad_norm": NaN,
-      "learning_rate": 9.493970231495835e-05,
+      "learning_rate": 0.0002498096996349117,
       "loss": 0.0,
       "step": 23
     },
     {
-      "epoch": 0.016233354470513633,
+      "epoch": 0.06174626145682586,
       "grad_norm": NaN,
-      "learning_rate": 9.414737964294636e-05,
+      "learning_rate": 0.0002452754788010787,
       "loss": 0.0,
       "step": 24
     },
     {
-      "epoch": 0.016909744240118367,
+      "epoch": 0.06431902235086026,
       "grad_norm": NaN,
-      "learning_rate": 9.330127018922194e-05,
+      "learning_rate": 0.00024059659612857536,
       "loss": 0.0,
       "step": 25
     },
     {
-      "epoch": 0.016909744240118367,
+      "epoch": 0.06431902235086026,
       "eval_loss": NaN,
-      "eval_runtime": 94.1534,
-      "eval_samples_per_second": 26.446,
-      "eval_steps_per_second": 13.223,
+      "eval_runtime": 1.4841,
+      "eval_samples_per_second": 33.691,
+      "eval_steps_per_second": 8.76,
       "step": 25
     },
     {
-      "epoch": 0.017586134009723104,
+      "epoch": 0.06689178324489468,
       "grad_norm": NaN,
-      "learning_rate": 9.24024048078213e-05,
+      "learning_rate": 0.00023578148328022626,
       "loss": 0.0,
       "step": 26
     },
     {
-      "epoch": 0.018262523779327838,
+      "epoch": 0.0694645441389291,
       "grad_norm": NaN,
-      "learning_rate": 9.145187862775209e-05,
+      "learning_rate": 0.00023083881741484068,
       "loss": 0.0,
       "step": 27
     },
     {
-      "epoch": 0.01893891354893257,
+      "epoch": 0.0720373050329635,
       "grad_norm": NaN,
-      "learning_rate": 9.045084971874738e-05,
+      "learning_rate": 0.00022577750555038587,
       "loss": 0.0,
       "step": 28
     },
     {
-      "epoch": 0.019615303318537308,
+      "epoch": 0.07461006592699791,
       "grad_norm": NaN,
-      "learning_rate": 8.940053768033609e-05,
+      "learning_rate": 0.000220606668512939,
       "loss": 0.0,
       "step": 29
     },
     {
-      "epoch": 0.020291693088142042,
+      "epoch": 0.07718282682103232,
       "grad_norm": NaN,
-      "learning_rate": 8.83022221559489e-05,
+      "learning_rate": 0.00021533562450034164,
       "loss": 0.0,
       "step": 30
     },
     {
-      "epoch": 0.020968082857746775,
+      "epoch": 0.07975558771506673,
       "grad_norm": NaN,
-      "learning_rate": 8.715724127386972e-05,
+      "learning_rate": 0.00020997387229017774,
       "loss": 0.0,
       "step": 31
     },
     {
-      "epoch": 0.021644472627351512,
+      "epoch": 0.08232834860910114,
       "grad_norm": NaN,
-      "learning_rate": 8.596699001693255e-05,
+      "learning_rate": 0.00020453107412233428,
       "loss": 0.0,
       "step": 32
     },
     {
-      "epoch": 0.022320862396956246,
+      "epoch": 0.08490110950313555,
       "grad_norm": NaN,
-      "learning_rate": 8.473291852294987e-05,
+      "learning_rate": 0.0001990170382869919,
       "loss": 0.0,
       "step": 33
     },
     {
-      "epoch": 0.02299725216656098,
+      "epoch": 0.08747387039716996,
       "grad_norm": NaN,
-      "learning_rate": 8.345653031794292e-05,
+      "learning_rate": 0.00019344170144942302,
       "loss": 0.0,
       "step": 34
     },
     {
-      "epoch": 0.023673641936165717,
+      "epoch": 0.09004663129120437,
       "grad_norm": NaN,
-      "learning_rate": 8.213938048432697e-05,
+      "learning_rate": 0.00018781511074344962,
       "loss": 0.0,
       "step": 35
     },
     {
-      "epoch": 0.02435003170577045,
+      "epoch": 0.09261939218523878,
       "grad_norm": NaN,
-      "learning_rate": 8.07830737662829e-05,
+      "learning_rate": 0.0001821474056658286,
       "loss": 0.0,
       "step": 36
     },
     {
-      "epoch": 0.025026421475375184,
+      "epoch": 0.0951921530792732,
       "grad_norm": NaN,
-      "learning_rate": 7.938926261462366e-05,
+      "learning_rate": 0.00017644879980419374,
       "loss": 0.0,
       "step": 37
     },
     {
-      "epoch": 0.02570281124497992,
+      "epoch": 0.09776491397330761,
       "grad_norm": NaN,
-      "learning_rate": 7.795964517353735e-05,
+      "learning_rate": 0.00017072956243148002,
       "loss": 0.0,
       "step": 38
     },
     {
-      "epoch": 0.026379201014584654,
+      "epoch": 0.10033767486734202,
       "grad_norm": NaN,
-      "learning_rate": 7.649596321166024e-05,
+      "learning_rate": 0.000165,
       "loss": 0.0,
       "step": 39
     },
     {
-      "epoch": 0.027055590784189388,
+      "epoch": 0.10291043576137643,
       "grad_norm": NaN,
-      "learning_rate": 7.500000000000001e-05,
+      "learning_rate": 0.00015927043756852,
       "loss": 0.0,
       "step": 40
     },
     {
-      "epoch": 0.027731980553794125,
+      "epoch": 0.10548319665541084,
       "grad_norm": NaN,
-      "learning_rate": 7.347357813929454e-05,
+      "learning_rate": 0.0001535512001958063,
       "loss": 0.0,
       "step": 41
     },
     {
-      "epoch": 0.02840837032339886,
+      "epoch": 0.10805595754944525,
       "grad_norm": NaN,
-      "learning_rate": 7.191855733945387e-05,
+      "learning_rate": 0.00014785259433417133,
       "loss": 0.0,
       "step": 42
     },
     {
-      "epoch": 0.029084760093003592,
+      "epoch": 0.11062871844347966,
       "grad_norm": NaN,
-      "learning_rate": 7.033683215379002e-05,
+      "learning_rate": 0.00014218488925655037,
       "loss": 0.0,
       "step": 43
     },
     {
-      "epoch": 0.02976114986260833,
+      "epoch": 0.11320147933751407,
       "grad_norm": NaN,
-      "learning_rate": 6.873032967079561e-05,
+      "learning_rate": 0.00013655829855057698,
       "loss": 0.0,
       "step": 44
     },
     {
-      "epoch": 0.030437539632213063,
+      "epoch": 0.11577424023154848,
       "grad_norm": NaN,
-      "learning_rate": 6.710100716628344e-05,
+      "learning_rate": 0.00013098296171300814,
       "loss": 0.0,
       "step": 45
     },
     {
-      "epoch": 0.031113929401817796,
+      "epoch": 0.11834700112558288,
       "grad_norm": NaN,
-      "learning_rate": 6.545084971874738e-05,
+      "learning_rate": 0.0001254689258776657,
       "loss": 0.0,
       "step": 46
     },
     {
-      "epoch": 0.03179031917142253,
+      "epoch": 0.12091976201961731,
       "grad_norm": NaN,
-      "learning_rate": 6.378186779084995e-05,
+      "learning_rate": 0.00012002612770982222,
       "loss": 0.0,
       "step": 47
     },
     {
-      "epoch": 0.03246670894102727,
+      "epoch": 0.12349252291365172,
       "grad_norm": NaN,
-      "learning_rate": 6.209609477998338e-05,
+      "learning_rate": 0.00011466437549965834,
       "loss": 0.0,
       "step": 48
     },
     {
-      "epoch": 0.033143098710632,
+      "epoch": 0.1260652838076861,
       "grad_norm": NaN,
-      "learning_rate": 6.0395584540887963e-05,
+      "learning_rate": 0.00010939333148706099,
       "loss": 0.0,
       "step": 49
     },
     {
-      "epoch": 0.033819488480236734,
+      "epoch": 0.12863804470172052,
       "grad_norm": NaN,
-      "learning_rate": 5.868240888334653e-05,
+      "learning_rate": 0.00010422249444961407,
       "loss": 0.0,
       "step": 50
     },
     {
-      "epoch": 0.033819488480236734,
+      "epoch": 0.12863804470172052,
       "eval_loss": NaN,
-      "eval_runtime": 94.0007,
-      "eval_samples_per_second": 26.489,
-      "eval_steps_per_second": 13.245,
+      "eval_runtime": 1.483,
+      "eval_samples_per_second": 33.716,
+      "eval_steps_per_second": 8.766,
       "step": 50
-    },
-    {
-      "epoch": 0.034495878249841475,
-      "grad_norm": NaN,
-      "learning_rate": 5.695865504800327e-05,
-      "loss": 0.0,
-      "step": 51
-    },
-    {
-      "epoch": 0.03517226801944621,
-      "grad_norm": NaN,
-      "learning_rate": 5.522642316338268e-05,
-      "loss": 0.0,
-      "step": 52
-    },
-    {
-      "epoch": 0.03584865778905094,
-      "grad_norm": NaN,
-      "learning_rate": 5.348782368720626e-05,
-      "loss": 0.0,
-      "step": 53
-    },
-    {
-      "epoch": 0.036525047558655675,
-      "grad_norm": NaN,
-      "learning_rate": 5.174497483512506e-05,
-      "loss": 0.0,
-      "step": 54
-    },
-    {
-      "epoch": 0.03720143732826041,
-      "grad_norm": NaN,
-      "learning_rate": 5e-05,
-      "loss": 0.0,
-      "step": 55
-    },
-    {
-      "epoch": 0.03787782709786514,
-      "grad_norm": NaN,
-      "learning_rate": 4.825502516487497e-05,
-      "loss": 0.0,
-      "step": 56
-    },
-    {
-      "epoch": 0.03855421686746988,
-      "grad_norm": NaN,
-      "learning_rate": 4.6512176312793736e-05,
-      "loss": 0.0,
-      "step": 57
-    },
-    {
-      "epoch": 0.039230606637074616,
-      "grad_norm": NaN,
-      "learning_rate": 4.477357683661734e-05,
-      "loss": 0.0,
-      "step": 58
-    },
-    {
-      "epoch": 0.03990699640667935,
-      "grad_norm": NaN,
-      "learning_rate": 4.3041344951996746e-05,
-      "loss": 0.0,
-      "step": 59
-    },
-    {
-      "epoch": 0.040583386176284084,
-      "grad_norm": NaN,
-      "learning_rate": 4.131759111665349e-05,
-      "loss": 0.0,
-      "step": 60
-    },
-    {
-      "epoch": 0.04125977594588882,
-      "grad_norm": NaN,
-      "learning_rate": 3.960441545911204e-05,
-      "loss": 0.0,
-      "step": 61
-    },
-    {
-      "epoch": 0.04193616571549355,
-      "grad_norm": NaN,
-      "learning_rate": 3.790390522001662e-05,
-      "loss": 0.0,
-      "step": 62
-    },
-    {
-      "epoch": 0.04261255548509829,
-      "grad_norm": NaN,
-      "learning_rate": 3.6218132209150045e-05,
-      "loss": 0.0,
-      "step": 63
-    },
-    {
-      "epoch": 0.043288945254703025,
-      "grad_norm": NaN,
-      "learning_rate": 3.4549150281252636e-05,
-      "loss": 0.0,
-      "step": 64
-    },
-    {
-      "epoch": 0.04396533502430776,
-      "grad_norm": NaN,
-      "learning_rate": 3.289899283371657e-05,
-      "loss": 0.0,
-      "step": 65
-    },
-    {
-      "epoch": 0.04464172479391249,
-      "grad_norm": NaN,
-      "learning_rate": 3.12696703292044e-05,
-      "loss": 0.0,
-      "step": 66
-    },
-    {
-      "epoch": 0.045318114563517226,
-      "grad_norm": NaN,
-      "learning_rate": 2.9663167846209998e-05,
-      "loss": 0.0,
-      "step": 67
-    },
-    {
-      "epoch": 0.04599450433312196,
-      "grad_norm": NaN,
-      "learning_rate": 2.8081442660546125e-05,
-      "loss": 0.0,
-      "step": 68
-    },
-    {
-      "epoch": 0.0466708941027267,
-      "grad_norm": NaN,
-      "learning_rate": 2.6526421860705473e-05,
-      "loss": 0.0,
-      "step": 69
-    },
-    {
-      "epoch": 0.04734728387233143,
-      "grad_norm": NaN,
-      "learning_rate": 2.500000000000001e-05,
-      "loss": 0.0,
-      "step": 70
-    },
-    {
-      "epoch": 0.04802367364193617,
-      "grad_norm": NaN,
-      "learning_rate": 2.350403678833976e-05,
-      "loss": 0.0,
-      "step": 71
-    },
-    {
-      "epoch": 0.0487000634115409,
-      "grad_norm": NaN,
-      "learning_rate": 2.2040354826462668e-05,
-      "loss": 0.0,
-      "step": 72
-    },
-    {
-      "epoch": 0.049376453181145634,
-      "grad_norm": NaN,
-      "learning_rate": 2.061073738537635e-05,
-      "loss": 0.0,
-      "step": 73
-    },
-    {
-      "epoch": 0.05005284295075037,
-      "grad_norm": NaN,
-      "learning_rate": 1.9216926233717085e-05,
-      "loss": 0.0,
-      "step": 74
-    },
-    {
-      "epoch": 0.05072923272035511,
-      "grad_norm": NaN,
-      "learning_rate": 1.7860619515673033e-05,
-      "loss": 0.0,
-      "step": 75
-    },
-    {
-      "epoch": 0.05072923272035511,
-      "eval_loss": NaN,
-      "eval_runtime": 104.1703,
-      "eval_samples_per_second": 23.903,
-      "eval_steps_per_second": 11.952,
-      "step": 75
-    },
-    {
-      "epoch": 0.05140562248995984,
-      "grad_norm": NaN,
-      "learning_rate": 1.6543469682057106e-05,
-      "loss": 0.0,
-      "step": 76
-    },
-    {
-      "epoch": 0.052082012259564575,
-      "grad_norm": NaN,
-      "learning_rate": 1.526708147705013e-05,
-      "loss": 0.0,
-      "step": 77
-    },
-    {
-      "epoch": 0.05275840202916931,
-      "grad_norm": NaN,
-      "learning_rate": 1.4033009983067452e-05,
-      "loss": 0.0,
-      "step": 78
-    },
-    {
-      "epoch": 0.05343479179877404,
-      "grad_norm": NaN,
-      "learning_rate": 1.2842758726130283e-05,
-      "loss": 0.0,
-      "step": 79
-    },
-    {
-      "epoch": 0.054111181568378776,
-      "grad_norm": NaN,
-      "learning_rate": 1.1697777844051105e-05,
-      "loss": 0.0,
-      "step": 80
-    },
-    {
-      "epoch": 0.054787571337983516,
-      "grad_norm": NaN,
-      "learning_rate": 1.0599462319663905e-05,
-      "loss": 0.0,
-      "step": 81
-    },
-    {
-      "epoch": 0.05546396110758825,
-      "grad_norm": NaN,
-      "learning_rate": 9.549150281252633e-06,
-      "loss": 0.0,
-      "step": 82
-    },
-    {
-      "epoch": 0.056140350877192984,
-      "grad_norm": NaN,
-      "learning_rate": 8.548121372247918e-06,
-      "loss": 0.0,
-      "step": 83
-    },
-    {
-      "epoch": 0.05681674064679772,
-      "grad_norm": NaN,
-      "learning_rate": 7.597595192178702e-06,
-      "loss": 0.0,
-      "step": 84
-    },
-    {
-      "epoch": 0.05749313041640245,
-      "grad_norm": NaN,
-      "learning_rate": 6.698729810778065e-06,
-      "loss": 0.0,
-      "step": 85
-    },
-    {
-      "epoch": 0.058169520186007184,
-      "grad_norm": NaN,
-      "learning_rate": 5.852620357053651e-06,
-      "loss": 0.0,
-      "step": 86
-    },
-    {
-      "epoch": 0.058845909955611925,
-      "grad_norm": NaN,
-      "learning_rate": 5.060297685041659e-06,
-      "loss": 0.0,
-      "step": 87
-    },
-    {
-      "epoch": 0.05952229972521666,
-      "grad_norm": NaN,
-      "learning_rate": 4.322727117869951e-06,
-      "loss": 0.0,
-      "step": 88
-    },
-    {
-      "epoch": 0.06019868949482139,
-      "grad_norm": NaN,
-      "learning_rate": 3.6408072716606346e-06,
-      "loss": 0.0,
-      "step": 89
-    },
-    {
-      "epoch": 0.060875079264426125,
-      "grad_norm": NaN,
-      "learning_rate": 3.0153689607045845e-06,
-      "loss": 0.0,
-      "step": 90
-    },
-    {
-      "epoch": 0.06155146903403086,
-      "grad_norm": NaN,
-      "learning_rate": 2.4471741852423237e-06,
-      "loss": 0.0,
-      "step": 91
-    },
-    {
-      "epoch": 0.06222785880363559,
-      "grad_norm": NaN,
-      "learning_rate": 1.9369152030840556e-06,
-      "loss": 0.0,
-      "step": 92
-    },
-    {
-      "epoch": 0.06290424857324033,
-      "grad_norm": NaN,
-      "learning_rate": 1.4852136862001764e-06,
-      "loss": 0.0,
-      "step": 93
-    },
-    {
-      "epoch": 0.06358063834284507,
-      "grad_norm": NaN,
-      "learning_rate": 1.0926199633097157e-06,
-      "loss": 0.0,
-      "step": 94
-    },
-    {
-      "epoch": 0.0642570281124498,
-      "grad_norm": NaN,
-      "learning_rate": 7.596123493895991e-07,
-      "loss": 0.0,
-      "step": 95
-    },
-    {
-      "epoch": 0.06493341788205453,
-      "grad_norm": NaN,
-      "learning_rate": 4.865965629214819e-07,
-      "loss": 0.0,
-      "step": 96
-    },
-    {
-      "epoch": 0.06560980765165927,
-      "grad_norm": NaN,
-      "learning_rate": 2.7390523158633554e-07,
-      "loss": 0.0,
-      "step": 97
-    },
-    {
-      "epoch": 0.066286197421264,
-      "grad_norm": NaN,
-      "learning_rate": 1.2179748700879012e-07,
-      "loss": 0.0,
-      "step": 98
-    },
-    {
-      "epoch": 0.06696258719086874,
-      "grad_norm": NaN,
-      "learning_rate": 3.04586490452119e-08,
-      "loss": 0.0,
-      "step": 99
-    },
-    {
-      "epoch": 0.06763897696047347,
-      "grad_norm": NaN,
-      "learning_rate": 0.0,
-      "loss": 0.0,
-      "step": 100
-    },
-    {
-      "epoch": 0.06763897696047347,
-      "eval_loss": NaN,
-      "eval_runtime": 195.2703,
-      "eval_samples_per_second": 12.752,
-      "eval_steps_per_second": 6.376,
-      "step": 100
     }
   ],
   "logging_steps": 1,
-  "max_steps": 100,
+  "max_steps": 76,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
-  "save_steps": 25,
+  "save_steps": 50,
   "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 1,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
     "TrainerControl": {
       "args": {
         "should_epoch_stop": false,
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": true
+        "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 1.402135828758528e+17,
-  "train_batch_size": 2,
+  "total_flos": 5.6659779262611456e+17,
+  "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null
 }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e8a6cbf1eff2ed6abcb6caa32099c42e7adfb184edf43841908c9d3350c9ec0
+oid sha256:012696984907ada12377abc123ba0f028211273cc21153dc2137425ed4139c0a
 size 6776

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e8a6cbf1eff2ed6abcb6caa32099c42e7adfb184edf43841908c9d3350c9ec0
+oid sha256:012696984907ada12377abc123ba0f028211273cc21153dc2137425ed4139c0a
 size 6776