Upload full trainer checkpoint step=1098 (profile=llama3_1_8b_base_lessdata, run_type=sft)

.gitattributes

@@ -150,3 +150,4 @@ meta-llama__meta-llama-3.1-8b/grpo/checkpoints/checkpoint-1013/tokenizer.json fi
 meta-llama__meta-llama-3.1-8b/grpo/checkpoints/checkpoint-2026/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-366/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-732/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098/tokenizer.json filter=lfs diff=lfs merge=lfs -text

meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098/README.md

@@ -0,0 +1,210 @@
+---
+base_model: unsloth/Meta-Llama-3.1-8B
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:unsloth/Meta-Llama-3.1-8B
+- lora
+- sft
+- transformers
+- trl
+- unsloth
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098/adapter_config.json

@@ -0,0 +1,50 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "LlamaForCausalLM",
+    "parent_library": "transformers.models.llama.modeling_llama",
+    "unsloth_fixed": true
+  },
+  "base_model_name_or_path": "unsloth/Meta-Llama-3.1-8B",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "up_proj",
+    "gate_proj",
+    "down_proj",
+    "o_proj",
+    "v_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3998cb16f83a5deb3ff6fe0e36e9c165afe76105855779789564d6663ece011e
+size 83945296

meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098/chat_template.jinja

@@ -0,0 +1,12 @@
+{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ messages[0]['content'] + '
+
+' }}{% set loop_messages = messages[1:] %}{% else %}{{ '' + '
+
+' }}{% set loop_messages = messages %}{% endif %}{% for message in loop_messages %}{% if message['role'] == 'user' %}{{ '### Instruction:
+' + message['content'] + '
+
+' }}{% elif message['role'] == 'assistant' %}{{ '### Response:
+' + message['content'] + eos_token + '
+
+' }}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '### Response:
+' }}{% endif %}

meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:167a8807a4f09ae154cc13e8241ad219ccc02ef35bfecd8cf5da4005845ac15d
+size 43127973

meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d569a9c96ff4e89d32dfe8ca0b97952cc2edcd76f89c14caa7e57238c4b46c8a
+size 14645

meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4812f164ea7757b51fbfb4d00b1fa6325746cb792d7c7ec0dd7e074068a93cfb
+size 1465

meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098/tokenizer_config.json

@@ -0,0 +1,18 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<|begin_of_text|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|end_of_text|>",
+  "from_slow": true,
+  "is_local": false,
+  "legacy": false,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 131072,
+  "pad_token": "<|finetune_right_pad_id|>",
+  "padding_side": "right",
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": null
+}

meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098/trainer_state.json

@@ -0,0 +1,1615 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.7489768076398363,
+  "eval_steps": 366,
+  "global_step": 1098,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0034106412005457027,
+      "grad_norm": 3.1571648120880127,
+      "learning_rate": 2.542372881355932e-06,
+      "loss": 1.6981744766235352,
+      "step": 5
+    },
+    {
+      "epoch": 0.0068212824010914054,
+      "grad_norm": 9.80073070526123,
+      "learning_rate": 5.720338983050847e-06,
+      "loss": 1.5801130294799806,
+      "step": 10
+    },
+    {
+      "epoch": 0.010231923601637109,
+      "grad_norm": 4.481558799743652,
+      "learning_rate": 8.898305084745763e-06,
+      "loss": 2.1718717575073243,
+      "step": 15
+    },
+    {
+      "epoch": 0.013642564802182811,
+      "grad_norm": 7.221553802490234,
+      "learning_rate": 1.2076271186440677e-05,
+      "loss": 1.5270480155944823,
+      "step": 20
+    },
+    {
+      "epoch": 0.017053206002728513,
+      "grad_norm": 5.330269813537598,
+      "learning_rate": 1.5254237288135592e-05,
+      "loss": 0.9586630821228027,
+      "step": 25
+    },
+    {
+      "epoch": 0.020463847203274217,
+      "grad_norm": 1.7404323816299438,
+      "learning_rate": 1.8432203389830506e-05,
+      "loss": 0.48505802154541017,
+      "step": 30
+    },
+    {
+      "epoch": 0.023874488403819918,
+      "grad_norm": 1.2418887615203857,
+      "learning_rate": 2.1610169491525424e-05,
+      "loss": 0.24452033042907714,
+      "step": 35
+    },
+    {
+      "epoch": 0.027285129604365622,
+      "grad_norm": 1.5928819179534912,
+      "learning_rate": 2.4788135593220338e-05,
+      "loss": 0.25337533950805663,
+      "step": 40
+    },
+    {
+      "epoch": 0.030695770804911322,
+      "grad_norm": 2.0335769653320312,
+      "learning_rate": 2.796610169491525e-05,
+      "loss": 0.1938886046409607,
+      "step": 45
+    },
+    {
+      "epoch": 0.034106412005457026,
+      "grad_norm": 0.18029411137104034,
+      "learning_rate": 3.114406779661017e-05,
+      "loss": 0.1834133505821228,
+      "step": 50
+    },
+    {
+      "epoch": 0.03751705320600273,
+      "grad_norm": 1.7757192850112915,
+      "learning_rate": 3.432203389830508e-05,
+      "loss": 0.15151114463806153,
+      "step": 55
+    },
+    {
+      "epoch": 0.040927694406548434,
+      "grad_norm": 2.1376893520355225,
+      "learning_rate": 3.75e-05,
+      "loss": 0.21634674072265625,
+      "step": 60
+    },
+    {
+      "epoch": 0.04433833560709413,
+      "grad_norm": 2.196387767791748,
+      "learning_rate": 4.067796610169491e-05,
+      "loss": 0.11320395469665527,
+      "step": 65
+    },
+    {
+      "epoch": 0.047748976807639835,
+      "grad_norm": 1.5888807773590088,
+      "learning_rate": 4.3855932203389825e-05,
+      "loss": 0.11050152778625488,
+      "step": 70
+    },
+    {
+      "epoch": 0.05115961800818554,
+      "grad_norm": 4.606944561004639,
+      "learning_rate": 4.703389830508474e-05,
+      "loss": 0.23255281448364257,
+      "step": 75
+    },
+    {
+      "epoch": 0.054570259208731244,
+      "grad_norm": 1.7308639287948608,
+      "learning_rate": 5.021186440677966e-05,
+      "loss": 0.08995866179466247,
+      "step": 80
+    },
+    {
+      "epoch": 0.05798090040927694,
+      "grad_norm": 2.622438430786133,
+      "learning_rate": 5.338983050847457e-05,
+      "loss": 0.05715223550796509,
+      "step": 85
+    },
+    {
+      "epoch": 0.061391541609822645,
+      "grad_norm": 0.34140461683273315,
+      "learning_rate": 5.656779661016949e-05,
+      "loss": 0.06703727245330811,
+      "step": 90
+    },
+    {
+      "epoch": 0.06480218281036836,
+      "grad_norm": 2.7264959812164307,
+      "learning_rate": 5.97457627118644e-05,
+      "loss": 0.10092716217041016,
+      "step": 95
+    },
+    {
+      "epoch": 0.06821282401091405,
+      "grad_norm": 2.0234780311584473,
+      "learning_rate": 6.292372881355932e-05,
+      "loss": 0.03579937815666199,
+      "step": 100
+    },
+    {
+      "epoch": 0.07162346521145975,
+      "grad_norm": 1.3716133832931519,
+      "learning_rate": 6.610169491525423e-05,
+      "loss": 0.07851418852806091,
+      "step": 105
+    },
+    {
+      "epoch": 0.07503410641200546,
+      "grad_norm": 1.2307227849960327,
+      "learning_rate": 6.927966101694914e-05,
+      "loss": 0.07370938658714295,
+      "step": 110
+    },
+    {
+      "epoch": 0.07844474761255116,
+      "grad_norm": 1.7142690420150757,
+      "learning_rate": 7.245762711864406e-05,
+      "loss": 0.08064572811126709,
+      "step": 115
+    },
+    {
+      "epoch": 0.08185538881309687,
+      "grad_norm": 1.343595266342163,
+      "learning_rate": 7.499999439507554e-05,
+      "loss": 0.11841402053833008,
+      "step": 120
+    },
+    {
+      "epoch": 0.08526603001364257,
+      "grad_norm": 1.2511327266693115,
+      "learning_rate": 7.499979822289558e-05,
+      "loss": 0.06974860429763793,
+      "step": 125
+    },
+    {
+      "epoch": 0.08867667121418826,
+      "grad_norm": 1.5642656087875366,
+      "learning_rate": 7.49993218061684e-05,
+      "loss": 0.10972714424133301,
+      "step": 130
+    },
+    {
+      "epoch": 0.09208731241473397,
+      "grad_norm": 1.6669789552688599,
+      "learning_rate": 7.499856514845436e-05,
+      "loss": 0.09864612817764282,
+      "step": 135
+    },
+    {
+      "epoch": 0.09549795361527967,
+      "grad_norm": 0.9789333343505859,
+      "learning_rate": 7.499752825540815e-05,
+      "loss": 0.0699621558189392,
+      "step": 140
+    },
+    {
+      "epoch": 0.09890859481582538,
+      "grad_norm": 0.41792476177215576,
+      "learning_rate": 7.499621113477873e-05,
+      "loss": 0.06734349727630615,
+      "step": 145
+    },
+    {
+      "epoch": 0.10231923601637108,
+      "grad_norm": 1.5968533754348755,
+      "learning_rate": 7.499461379640919e-05,
+      "loss": 0.060729533433914185,
+      "step": 150
+    },
+    {
+      "epoch": 0.10572987721691678,
+      "grad_norm": 0.8512193560600281,
+      "learning_rate": 7.499273625223683e-05,
+      "loss": 0.052730172872543335,
+      "step": 155
+    },
+    {
+      "epoch": 0.10914051841746249,
+      "grad_norm": 1.3257211446762085,
+      "learning_rate": 7.499057851629299e-05,
+      "loss": 0.10094538927078248,
+      "step": 160
+    },
+    {
+      "epoch": 0.11255115961800818,
+      "grad_norm": 0.659118115901947,
+      "learning_rate": 7.498814060470288e-05,
+      "loss": 0.01604635864496231,
+      "step": 165
+    },
+    {
+      "epoch": 0.11596180081855388,
+      "grad_norm": 2.454979181289673,
+      "learning_rate": 7.49854225356856e-05,
+      "loss": 0.13272385597229003,
+      "step": 170
+    },
+    {
+      "epoch": 0.11937244201909959,
+      "grad_norm": 2.510040521621704,
+      "learning_rate": 7.498242432955388e-05,
+      "loss": 0.08396227955818177,
+      "step": 175
+    },
+    {
+      "epoch": 0.12278308321964529,
+      "grad_norm": 0.2637259364128113,
+      "learning_rate": 7.4979146008714e-05,
+      "loss": 0.05852065086364746,
+      "step": 180
+    },
+    {
+      "epoch": 0.126193724420191,
+      "grad_norm": 1.4320851564407349,
+      "learning_rate": 7.497558759766564e-05,
+      "loss": 0.10392802953720093,
+      "step": 185
+    },
+    {
+      "epoch": 0.1296043656207367,
+      "grad_norm": 1.273027777671814,
+      "learning_rate": 7.497174912300156e-05,
+      "loss": 0.08062989711761474,
+      "step": 190
+    },
+    {
+      "epoch": 0.1330150068212824,
+      "grad_norm": 0.8102871179580688,
+      "learning_rate": 7.496763061340759e-05,
+      "loss": 0.07294153571128845,
+      "step": 195
+    },
+    {
+      "epoch": 0.1364256480218281,
+      "grad_norm": 1.951328992843628,
+      "learning_rate": 7.496323209966228e-05,
+      "loss": 0.07738351821899414,
+      "step": 200
+    },
+    {
+      "epoch": 0.13983628922237382,
+      "grad_norm": 0.3880983889102936,
+      "learning_rate": 7.495855361463674e-05,
+      "loss": 0.07225048542022705,
+      "step": 205
+    },
+    {
+      "epoch": 0.1432469304229195,
+      "grad_norm": 3.3205058574676514,
+      "learning_rate": 7.495359519329433e-05,
+      "loss": 0.05682974457740784,
+      "step": 210
+    },
+    {
+      "epoch": 0.1466575716234652,
+      "grad_norm": 0.9203559160232544,
+      "learning_rate": 7.49483568726905e-05,
+      "loss": 0.08767472505569458,
+      "step": 215
+    },
+    {
+      "epoch": 0.15006821282401092,
+      "grad_norm": 0.585360586643219,
+      "learning_rate": 7.494283869197239e-05,
+      "loss": 0.039227068424224854,
+      "step": 220
+    },
+    {
+      "epoch": 0.1534788540245566,
+      "grad_norm": 1.7096059322357178,
+      "learning_rate": 7.493704069237862e-05,
+      "loss": 0.10281096696853638,
+      "step": 225
+    },
+    {
+      "epoch": 0.15688949522510232,
+      "grad_norm": 0.4110204875469208,
+      "learning_rate": 7.493096291723898e-05,
+      "loss": 0.04346161186695099,
+      "step": 230
+    },
+    {
+      "epoch": 0.16030013642564803,
+      "grad_norm": 1.3272292613983154,
+      "learning_rate": 7.492460541197404e-05,
+      "loss": 0.049719154834747314,
+      "step": 235
+    },
+    {
+      "epoch": 0.16371077762619374,
+      "grad_norm": 1.1005016565322876,
+      "learning_rate": 7.491796822409494e-05,
+      "loss": 0.09335108399391175,
+      "step": 240
+    },
+    {
+      "epoch": 0.16712141882673942,
+      "grad_norm": 0.7811501026153564,
+      "learning_rate": 7.491105140320285e-05,
+      "loss": 0.05943926572799683,
+      "step": 245
+    },
+    {
+      "epoch": 0.17053206002728513,
+      "grad_norm": 1.4607417583465576,
+      "learning_rate": 7.490385500098879e-05,
+      "loss": 0.04385361075401306,
+      "step": 250
+    },
+    {
+      "epoch": 0.17394270122783084,
+      "grad_norm": 0.394960880279541,
+      "learning_rate": 7.489637907123308e-05,
+      "loss": 0.04446137547492981,
+      "step": 255
+    },
+    {
+      "epoch": 0.17735334242837653,
+      "grad_norm": 0.8768635988235474,
+      "learning_rate": 7.488862366980505e-05,
+      "loss": 0.04143576025962829,
+      "step": 260
+    },
+    {
+      "epoch": 0.18076398362892224,
+      "grad_norm": 1.9996010065078735,
+      "learning_rate": 7.488058885466262e-05,
+      "loss": 0.07952215671539306,
+      "step": 265
+    },
+    {
+      "epoch": 0.18417462482946795,
+      "grad_norm": 0.03770223259925842,
+      "learning_rate": 7.487227468585178e-05,
+      "loss": 0.02531362771987915,
+      "step": 270
+    },
+    {
+      "epoch": 0.18758526603001363,
+      "grad_norm": 0.26082542538642883,
+      "learning_rate": 7.486368122550619e-05,
+      "loss": 0.09930967688560485,
+      "step": 275
+    },
+    {
+      "epoch": 0.19099590723055934,
+      "grad_norm": 5.622270584106445,
+      "learning_rate": 7.485480853784677e-05,
+      "loss": 0.06534865498542786,
+      "step": 280
+    },
+    {
+      "epoch": 0.19440654843110505,
+      "grad_norm": 0.5298851132392883,
+      "learning_rate": 7.484565668918111e-05,
+      "loss": 0.06109699010848999,
+      "step": 285
+    },
+    {
+      "epoch": 0.19781718963165076,
+      "grad_norm": 1.4887421131134033,
+      "learning_rate": 7.483622574790308e-05,
+      "loss": 0.048966211080551145,
+      "step": 290
+    },
+    {
+      "epoch": 0.20122783083219645,
+      "grad_norm": 0.5699282884597778,
+      "learning_rate": 7.482651578449223e-05,
+      "loss": 0.05427658557891846,
+      "step": 295
+    },
+    {
+      "epoch": 0.20463847203274216,
+      "grad_norm": 1.6645292043685913,
+      "learning_rate": 7.481652687151339e-05,
+      "loss": 0.037466832995414735,
+      "step": 300
+    },
+    {
+      "epoch": 0.20804911323328787,
+      "grad_norm": 0.4979431629180908,
+      "learning_rate": 7.480625908361593e-05,
+      "loss": 0.019084173440933227,
+      "step": 305
+    },
+    {
+      "epoch": 0.21145975443383355,
+      "grad_norm": 2.73081636428833,
+      "learning_rate": 7.479571249753339e-05,
+      "loss": 0.07597044706344605,
+      "step": 310
+    },
+    {
+      "epoch": 0.21487039563437926,
+      "grad_norm": 0.009097559377551079,
+      "learning_rate": 7.478488719208281e-05,
+      "loss": 0.017771795392036438,
+      "step": 315
+    },
+    {
+      "epoch": 0.21828103683492497,
+      "grad_norm": 1.5284112691879272,
+      "learning_rate": 7.477378324816419e-05,
+      "loss": 0.07524526119232178,
+      "step": 320
+    },
+    {
+      "epoch": 0.22169167803547066,
+      "grad_norm": 1.400959849357605,
+      "learning_rate": 7.47624007487598e-05,
+      "loss": 0.0357323557138443,
+      "step": 325
+    },
+    {
+      "epoch": 0.22510231923601637,
+      "grad_norm": 0.5988397598266602,
+      "learning_rate": 7.47507397789337e-05,
+      "loss": 0.06072888970375061,
+      "step": 330
+    },
+    {
+      "epoch": 0.22851296043656208,
+      "grad_norm": 0.18309183418750763,
+      "learning_rate": 7.473880042583092e-05,
+      "loss": 0.03904334008693695,
+      "step": 335
+    },
+    {
+      "epoch": 0.23192360163710776,
+      "grad_norm": 0.7360084056854248,
+      "learning_rate": 7.472658277867702e-05,
+      "loss": 0.05045387148857117,
+      "step": 340
+    },
+    {
+      "epoch": 0.23533424283765347,
+      "grad_norm": 2.315072536468506,
+      "learning_rate": 7.471408692877724e-05,
+      "loss": 0.07920202016830444,
+      "step": 345
+    },
+    {
+      "epoch": 0.23874488403819918,
+      "grad_norm": 1.2811086177825928,
+      "learning_rate": 7.470131296951592e-05,
+      "loss": 0.05552580952644348,
+      "step": 350
+    },
+    {
+      "epoch": 0.2421555252387449,
+      "grad_norm": 4.006563186645508,
+      "learning_rate": 7.468826099635578e-05,
+      "loss": 0.1419215679168701,
+      "step": 355
+    },
+    {
+      "epoch": 0.24556616643929058,
+      "grad_norm": 1.1540688276290894,
+      "learning_rate": 7.467493110683718e-05,
+      "loss": 0.03980849981307984,
+      "step": 360
+    },
+    {
+      "epoch": 0.2489768076398363,
+      "grad_norm": 1.5472272634506226,
+      "learning_rate": 7.466132340057742e-05,
+      "loss": 0.020862475037574768,
+      "step": 365
+    },
+    {
+      "epoch": 0.24965893587994542,
+      "eval_loss": 0.11654457449913025,
+      "eval_runtime": 1.0333,
+      "eval_samples_per_second": 72.584,
+      "eval_steps_per_second": 1.936,
+      "step": 366
+    },
+    {
+      "eval_cer_subset": 0.05232320479629377,
+      "eval_cer_subset_edit_distance": 384,
+      "eval_cer_subset_groups": 75,
+      "eval_cer_subset_items": 75,
+      "eval_cer_subset_ref_chars": 7339,
+      "step": 366
+    },
+    {
+      "epoch": 0.252387448840382,
+      "grad_norm": 1.730627417564392,
+      "learning_rate": 7.464743797927002e-05,
+      "loss": 0.11239330768585205,
+      "step": 370
+    },
+    {
+      "epoch": 0.2557980900409277,
+      "grad_norm": 0.1921682506799698,
+      "learning_rate": 7.463327494668388e-05,
+      "loss": 0.09260941743850708,
+      "step": 375
+    },
+    {
+      "epoch": 0.2592087312414734,
+      "grad_norm": 1.9259151220321655,
+      "learning_rate": 7.461883440866259e-05,
+      "loss": 0.03999299705028534,
+      "step": 380
+    },
+    {
+      "epoch": 0.2626193724420191,
+      "grad_norm": 0.0488249845802784,
+      "learning_rate": 7.460411647312358e-05,
+      "loss": 0.01459498256444931,
+      "step": 385
+    },
+    {
+      "epoch": 0.2660300136425648,
+      "grad_norm": 1.1967735290527344,
+      "learning_rate": 7.458912125005732e-05,
+      "loss": 0.17716412544250487,
+      "step": 390
+    },
+    {
+      "epoch": 0.2694406548431105,
+      "grad_norm": 1.0083205699920654,
+      "learning_rate": 7.457384885152655e-05,
+      "loss": 0.08738511800765991,
+      "step": 395
+    },
+    {
+      "epoch": 0.2728512960436562,
+      "grad_norm": 0.6705593466758728,
+      "learning_rate": 7.455829939166539e-05,
+      "loss": 0.026945650577545166,
+      "step": 400
+    },
+    {
+      "epoch": 0.2762619372442019,
+      "grad_norm": 1.0791361331939697,
+      "learning_rate": 7.45424729866785e-05,
+      "loss": 0.03804046213626862,
+      "step": 405
+    },
+    {
+      "epoch": 0.27967257844474763,
+      "grad_norm": 0.7377910017967224,
+      "learning_rate": 7.452636975484021e-05,
+      "loss": 0.0464675635099411,
+      "step": 410
+    },
+    {
+      "epoch": 0.2830832196452933,
+      "grad_norm": 0.8061625957489014,
+      "learning_rate": 7.450998981649365e-05,
+      "loss": 0.02737331986427307,
+      "step": 415
+    },
+    {
+      "epoch": 0.286493860845839,
+      "grad_norm": 0.2580685019493103,
+      "learning_rate": 7.449333329404982e-05,
+      "loss": 0.018785761296749116,
+      "step": 420
+    },
+    {
+      "epoch": 0.28990450204638474,
+      "grad_norm": 0.017052194103598595,
+      "learning_rate": 7.447640031198675e-05,
+      "loss": 0.11424320936203003,
+      "step": 425
+    },
+    {
+      "epoch": 0.2933151432469304,
+      "grad_norm": 0.2855948805809021,
+      "learning_rate": 7.445919099684845e-05,
+      "loss": 0.012821969389915467,
+      "step": 430
+    },
+    {
+      "epoch": 0.2967257844474761,
+      "grad_norm": 1.5070558786392212,
+      "learning_rate": 7.444170547724405e-05,
+      "loss": 0.037783479690551756,
+      "step": 435
+    },
+    {
+      "epoch": 0.30013642564802184,
+      "grad_norm": 0.18241967260837555,
+      "learning_rate": 7.442394388384684e-05,
+      "loss": 0.03347713351249695,
+      "step": 440
+    },
+    {
+      "epoch": 0.3035470668485675,
+      "grad_norm": 0.37319424748420715,
+      "learning_rate": 7.440590634939327e-05,
+      "loss": 0.05499382615089417,
+      "step": 445
+    },
+    {
+      "epoch": 0.3069577080491132,
+      "grad_norm": 0.11083097755908966,
+      "learning_rate": 7.438759300868193e-05,
+      "loss": 0.021977408230304717,
+      "step": 450
+    },
+    {
+      "epoch": 0.31036834924965895,
+      "grad_norm": 0.24985608458518982,
+      "learning_rate": 7.436900399857261e-05,
+      "loss": 0.07826730608940125,
+      "step": 455
+    },
+    {
+      "epoch": 0.31377899045020463,
+      "grad_norm": 2.5360186100006104,
+      "learning_rate": 7.43501394579852e-05,
+      "loss": 0.080865478515625,
+      "step": 460
+    },
+    {
+      "epoch": 0.3171896316507503,
+      "grad_norm": 0.7053658366203308,
+      "learning_rate": 7.433099952789876e-05,
+      "loss": 0.012464526295661926,
+      "step": 465
+    },
+    {
+      "epoch": 0.32060027285129605,
+      "grad_norm": 0.3297032117843628,
+      "learning_rate": 7.43115843513503e-05,
+      "loss": 0.05623521208763123,
+      "step": 470
+    },
+    {
+      "epoch": 0.32401091405184174,
+      "grad_norm": 1.3899471759796143,
+      "learning_rate": 7.42918940734339e-05,
+      "loss": 0.07306370735168458,
+      "step": 475
+    },
+    {
+      "epoch": 0.3274215552523875,
+      "grad_norm": 0.9437380433082581,
+      "learning_rate": 7.427192884129948e-05,
+      "loss": 0.058290761709213254,
+      "step": 480
+    },
+    {
+      "epoch": 0.33083219645293316,
+      "grad_norm": 1.8157323598861694,
+      "learning_rate": 7.42516888041518e-05,
+      "loss": 0.058532989025115965,
+      "step": 485
+    },
+    {
+      "epoch": 0.33424283765347884,
+      "grad_norm": 0.6275774836540222,
+      "learning_rate": 7.423117411324924e-05,
+      "loss": 0.0624964714050293,
+      "step": 490
+    },
+    {
+      "epoch": 0.3376534788540246,
+      "grad_norm": 0.6674565672874451,
+      "learning_rate": 7.421038492190278e-05,
+      "loss": 0.020014175772666933,
+      "step": 495
+    },
+    {
+      "epoch": 0.34106412005457026,
+      "grad_norm": 0.6229842901229858,
+      "learning_rate": 7.418932138547481e-05,
+      "loss": 0.03286575376987457,
+      "step": 500
+    },
+    {
+      "epoch": 0.34447476125511595,
+      "grad_norm": 1.441677212715149,
+      "learning_rate": 7.41679836613779e-05,
+      "loss": 0.04557921886444092,
+      "step": 505
+    },
+    {
+      "epoch": 0.3478854024556617,
+      "grad_norm": 0.8439592719078064,
+      "learning_rate": 7.414637190907379e-05,
+      "loss": 0.027792316675186158,
+      "step": 510
+    },
+    {
+      "epoch": 0.35129604365620737,
+      "grad_norm": 0.1357346475124359,
+      "learning_rate": 7.412448629007198e-05,
+      "loss": 0.024153730273246764,
+      "step": 515
+    },
+    {
+      "epoch": 0.35470668485675305,
+      "grad_norm": 0.11876281350851059,
+      "learning_rate": 7.41023269679287e-05,
+      "loss": 0.11651531457901002,
+      "step": 520
+    },
+    {
+      "epoch": 0.3581173260572988,
+      "grad_norm": 0.8576210737228394,
+      "learning_rate": 7.407989410824566e-05,
+      "loss": 0.045156928896903994,
+      "step": 525
+    },
+    {
+      "epoch": 0.3615279672578445,
+      "grad_norm": 0.39947113394737244,
+      "learning_rate": 7.40571878786687e-05,
+      "loss": 0.02562606930732727,
+      "step": 530
+    },
+    {
+      "epoch": 0.36493860845839016,
+      "grad_norm": 0.8822716474533081,
+      "learning_rate": 7.403420844888668e-05,
+      "loss": 0.05394383668899536,
+      "step": 535
+    },
+    {
+      "epoch": 0.3683492496589359,
+      "grad_norm": 1.8026832342147827,
+      "learning_rate": 7.40109559906301e-05,
+      "loss": 0.07706952095031738,
+      "step": 540
+    },
+    {
+      "epoch": 0.3717598908594816,
+      "grad_norm": 0.28902706503868103,
+      "learning_rate": 7.398743067766987e-05,
+      "loss": 0.0352792352437973,
+      "step": 545
+    },
+    {
+      "epoch": 0.37517053206002726,
+      "grad_norm": 0.2759908437728882,
+      "learning_rate": 7.396363268581609e-05,
+      "loss": 0.038266700506210324,
+      "step": 550
+    },
+    {
+      "epoch": 0.378581173260573,
+      "grad_norm": 1.0520153045654297,
+      "learning_rate": 7.39395621929165e-05,
+      "loss": 0.04206843376159668,
+      "step": 555
+    },
+    {
+      "epoch": 0.3819918144611187,
+      "grad_norm": 0.29290419816970825,
+      "learning_rate": 7.391521937885543e-05,
+      "loss": 0.04060278534889221,
+      "step": 560
+    },
+    {
+      "epoch": 0.38540245566166437,
+      "grad_norm": 0.11243477463722229,
+      "learning_rate": 7.389060442555228e-05,
+      "loss": 0.05412468910217285,
+      "step": 565
+    },
+    {
+      "epoch": 0.3888130968622101,
+      "grad_norm": 1.9416879415512085,
+      "learning_rate": 7.386571751696019e-05,
+      "loss": 0.02231921851634979,
+      "step": 570
+    },
+    {
+      "epoch": 0.3922237380627558,
+      "grad_norm": 0.5937671661376953,
+      "learning_rate": 7.384055883906474e-05,
+      "loss": 0.032561862468719484,
+      "step": 575
+    },
+    {
+      "epoch": 0.3956343792633015,
+      "grad_norm": 0.026148535311222076,
+      "learning_rate": 7.381512857988244e-05,
+      "loss": 0.07647547125816345,
+      "step": 580
+    },
+    {
+      "epoch": 0.3990450204638472,
+      "grad_norm": 0.7875772714614868,
+      "learning_rate": 7.378942692945944e-05,
+      "loss": 0.031203645467758178,
+      "step": 585
+    },
+    {
+      "epoch": 0.4024556616643929,
+      "grad_norm": 0.45512810349464417,
+      "learning_rate": 7.376345407987002e-05,
+      "loss": 0.04238590002059937,
+      "step": 590
+    },
+    {
+      "epoch": 0.40586630286493863,
+      "grad_norm": 1.66355299949646,
+      "learning_rate": 7.373721022521521e-05,
+      "loss": 0.052533066272735594,
+      "step": 595
+    },
+    {
+      "epoch": 0.4092769440654843,
+      "grad_norm": 0.08107655495405197,
+      "learning_rate": 7.371069556162133e-05,
+      "loss": 0.017715978622436523,
+      "step": 600
+    },
+    {
+      "epoch": 0.41268758526603,
+      "grad_norm": 0.32274800539016724,
+      "learning_rate": 7.368391028723851e-05,
+      "loss": 0.1379294991493225,
+      "step": 605
+    },
+    {
+      "epoch": 0.41609822646657574,
+      "grad_norm": 1.8197475671768188,
+      "learning_rate": 7.365685460223922e-05,
+      "loss": 0.03312918543815613,
+      "step": 610
+    },
+    {
+      "epoch": 0.4195088676671214,
+      "grad_norm": 0.1390945166349411,
+      "learning_rate": 7.362952870881677e-05,
+      "loss": 0.027584537863731384,
+      "step": 615
+    },
+    {
+      "epoch": 0.4229195088676671,
+      "grad_norm": 0.9081276655197144,
+      "learning_rate": 7.360193281118378e-05,
+      "loss": 0.06143233776092529,
+      "step": 620
+    },
+    {
+      "epoch": 0.42633015006821284,
+      "grad_norm": 0.07777975499629974,
+      "learning_rate": 7.35740671155707e-05,
+      "loss": 0.053598570823669436,
+      "step": 625
+    },
+    {
+      "epoch": 0.4297407912687585,
+      "grad_norm": 0.9314269423484802,
+      "learning_rate": 7.354593183022422e-05,
+      "loss": 0.05946495532989502,
+      "step": 630
+    },
+    {
+      "epoch": 0.4331514324693042,
+      "grad_norm": 0.5312000513076782,
+      "learning_rate": 7.351752716540575e-05,
+      "loss": 0.030707958340644836,
+      "step": 635
+    },
+    {
+      "epoch": 0.43656207366984995,
+      "grad_norm": 0.855117917060852,
+      "learning_rate": 7.348885333338984e-05,
+      "loss": 0.09321808815002441,
+      "step": 640
+    },
+    {
+      "epoch": 0.43997271487039563,
+      "grad_norm": 0.12914253771305084,
+      "learning_rate": 7.345991054846257e-05,
+      "loss": 0.010356919467449188,
+      "step": 645
+    },
+    {
+      "epoch": 0.4433833560709413,
+      "grad_norm": 0.4129096567630768,
+      "learning_rate": 7.343069902691999e-05,
+      "loss": 0.054264682531356814,
+      "step": 650
+    },
+    {
+      "epoch": 0.44679399727148705,
+      "grad_norm": 1.8499324321746826,
+      "learning_rate": 7.340121898706643e-05,
+      "loss": 0.050659948587417604,
+      "step": 655
+    },
+    {
+      "epoch": 0.45020463847203274,
+      "grad_norm": 0.5490806698799133,
+      "learning_rate": 7.337147064921299e-05,
+      "loss": 0.07158003449440002,
+      "step": 660
+    },
+    {
+      "epoch": 0.4536152796725784,
+      "grad_norm": 1.1408376693725586,
+      "learning_rate": 7.334145423567575e-05,
+      "loss": 0.08845412135124206,
+      "step": 665
+    },
+    {
+      "epoch": 0.45702592087312416,
+      "grad_norm": 1.5242546796798706,
+      "learning_rate": 7.331116997077426e-05,
+      "loss": 0.07773985266685486,
+      "step": 670
+    },
+    {
+      "epoch": 0.46043656207366984,
+      "grad_norm": 0.7061560153961182,
+      "learning_rate": 7.32806180808297e-05,
+      "loss": 0.047228410840034485,
+      "step": 675
+    },
+    {
+      "epoch": 0.4638472032742155,
+      "grad_norm": 0.8088539838790894,
+      "learning_rate": 7.324979879416333e-05,
+      "loss": 0.03726888597011566,
+      "step": 680
+    },
+    {
+      "epoch": 0.46725784447476126,
+      "grad_norm": 0.5670620799064636,
+      "learning_rate": 7.321871234109472e-05,
+      "loss": 0.02899191677570343,
+      "step": 685
+    },
+    {
+      "epoch": 0.47066848567530695,
+      "grad_norm": 2.3821427822113037,
+      "learning_rate": 7.318735895394e-05,
+      "loss": 0.033483856916427614,
+      "step": 690
+    },
+    {
+      "epoch": 0.4740791268758527,
+      "grad_norm": 0.8073883652687073,
+      "learning_rate": 7.315573886701023e-05,
+      "loss": 0.05756385326385498,
+      "step": 695
+    },
+    {
+      "epoch": 0.47748976807639837,
+      "grad_norm": 0.09120920300483704,
+      "learning_rate": 7.31238523166095e-05,
+      "loss": 0.0338085800409317,
+      "step": 700
+    },
+    {
+      "epoch": 0.48090040927694405,
+      "grad_norm": 0.33443862199783325,
+      "learning_rate": 7.309169954103326e-05,
+      "loss": 0.00844155102968216,
+      "step": 705
+    },
+    {
+      "epoch": 0.4843110504774898,
+      "grad_norm": 0.4880702793598175,
+      "learning_rate": 7.305928078056657e-05,
+      "loss": 0.09532383680343628,
+      "step": 710
+    },
+    {
+      "epoch": 0.4877216916780355,
+      "grad_norm": 0.11862733215093613,
+      "learning_rate": 7.302659627748221e-05,
+      "loss": 0.01845739334821701,
+      "step": 715
+    },
+    {
+      "epoch": 0.49113233287858116,
+      "grad_norm": 0.03655651956796646,
+      "learning_rate": 7.299364627603892e-05,
+      "loss": 0.030477851629257202,
+      "step": 720
+    },
+    {
+      "epoch": 0.4945429740791269,
+      "grad_norm": 1.481441617012024,
+      "learning_rate": 7.29604310224796e-05,
+      "loss": 0.07586092352867127,
+      "step": 725
+    },
+    {
+      "epoch": 0.4979536152796726,
+      "grad_norm": 0.8510580658912659,
+      "learning_rate": 7.292695076502938e-05,
+      "loss": 0.03589251637458801,
+      "step": 730
+    },
+    {
+      "epoch": 0.49931787175989084,
+      "eval_loss": 0.061700768768787384,
+      "eval_runtime": 0.8886,
+      "eval_samples_per_second": 84.399,
+      "eval_steps_per_second": 2.251,
+      "step": 732
+    },
+    {
+      "eval_cer_subset": 0.021256301948494344,
+      "eval_cer_subset_edit_distance": 156,
+      "eval_cer_subset_groups": 75,
+      "eval_cer_subset_items": 75,
+      "eval_cer_subset_ref_chars": 7339,
+      "step": 732
+    },
+    {
+      "epoch": 0.5013642564802183,
+      "grad_norm": 0.49610504508018494,
+      "learning_rate": 7.28932057538939e-05,
+      "loss": 0.06440846920013428,
+      "step": 735
+    },
+    {
+      "epoch": 0.504774897680764,
+      "grad_norm": 0.5659550428390503,
+      "learning_rate": 7.285919624125732e-05,
+      "loss": 0.08426347374916077,
+      "step": 740
+    },
+    {
+      "epoch": 0.5081855388813097,
+      "grad_norm": 0.04723689705133438,
+      "learning_rate": 7.282492248128047e-05,
+      "loss": 0.07788341641426086,
+      "step": 745
+    },
+    {
+      "epoch": 0.5115961800818554,
+      "grad_norm": 0.720941960811615,
+      "learning_rate": 7.2790384730099e-05,
+      "loss": 0.03100808262825012,
+      "step": 750
+    },
+    {
+      "epoch": 0.515006821282401,
+      "grad_norm": 0.652988851070404,
+      "learning_rate": 7.275558324582138e-05,
+      "loss": 0.03954651951789856,
+      "step": 755
+    },
+    {
+      "epoch": 0.5184174624829468,
+      "grad_norm": 1.2214330434799194,
+      "learning_rate": 7.272051828852705e-05,
+      "loss": 0.019992084801197053,
+      "step": 760
+    },
+    {
+      "epoch": 0.5218281036834925,
+      "grad_norm": 1.292953372001648,
+      "learning_rate": 7.268519012026443e-05,
+      "loss": 0.07394988536834717,
+      "step": 765
+    },
+    {
+      "epoch": 0.5252387448840382,
+      "grad_norm": 1.1823278665542603,
+      "learning_rate": 7.264959900504901e-05,
+      "loss": 0.037449967861175534,
+      "step": 770
+    },
+    {
+      "epoch": 0.5286493860845839,
+      "grad_norm": 1.1314970254898071,
+      "learning_rate": 7.261374520886128e-05,
+      "loss": 0.04381995797157288,
+      "step": 775
+    },
+    {
+      "epoch": 0.5320600272851296,
+      "grad_norm": 0.02543286792933941,
+      "learning_rate": 7.257762899964486e-05,
+      "loss": 0.07130052447319031,
+      "step": 780
+    },
+    {
+      "epoch": 0.5354706684856753,
+      "grad_norm": 0.37440457940101624,
+      "learning_rate": 7.25412506473044e-05,
+      "loss": 0.050841158628463744,
+      "step": 785
+    },
+    {
+      "epoch": 0.538881309686221,
+      "grad_norm": 0.2532084882259369,
+      "learning_rate": 7.250461042370365e-05,
+      "loss": 0.03486245274543762,
+      "step": 790
+    },
+    {
+      "epoch": 0.5422919508867667,
+      "grad_norm": 1.0150209665298462,
+      "learning_rate": 7.246770860266333e-05,
+      "loss": 0.050749993324279784,
+      "step": 795
+    },
+    {
+      "epoch": 0.5457025920873124,
+      "grad_norm": 1.108716607093811,
+      "learning_rate": 7.24305454599592e-05,
+      "loss": 0.03166365325450897,
+      "step": 800
+    },
+    {
+      "epoch": 0.5491132332878581,
+      "grad_norm": 0.16657976806163788,
+      "learning_rate": 7.239312127331989e-05,
+      "loss": 0.05016656517982483,
+      "step": 805
+    },
+    {
+      "epoch": 0.5525238744884038,
+      "grad_norm": 0.005627671722322702,
+      "learning_rate": 7.235543632242488e-05,
+      "loss": 0.021701858937740327,
+      "step": 810
+    },
+    {
+      "epoch": 0.5559345156889495,
+      "grad_norm": 1.8796381950378418,
+      "learning_rate": 7.231749088890241e-05,
+      "loss": 0.061094462871551514,
+      "step": 815
+    },
+    {
+      "epoch": 0.5593451568894953,
+      "grad_norm": 0.020010950043797493,
+      "learning_rate": 7.227928525632737e-05,
+      "loss": 0.0238655224442482,
+      "step": 820
+    },
+    {
+      "epoch": 0.562755798090041,
+      "grad_norm": 1.9777793884277344,
+      "learning_rate": 7.224081971021914e-05,
+      "loss": 0.041665592789649965,
+      "step": 825
+    },
+    {
+      "epoch": 0.5661664392905866,
+      "grad_norm": 0.06644955277442932,
+      "learning_rate": 7.220209453803954e-05,
+      "loss": 0.016651667654514313,
+      "step": 830
+    },
+    {
+      "epoch": 0.5695770804911323,
+      "grad_norm": 0.8203716278076172,
+      "learning_rate": 7.216311002919064e-05,
+      "loss": 0.03519523441791535,
+      "step": 835
+    },
+    {
+      "epoch": 0.572987721691678,
+      "grad_norm": 1.957996129989624,
+      "learning_rate": 7.212386647501254e-05,
+      "loss": 0.03704521656036377,
+      "step": 840
+    },
+    {
+      "epoch": 0.5763983628922238,
+      "grad_norm": 0.2386065572500229,
+      "learning_rate": 7.208436416878125e-05,
+      "loss": 0.02845575213432312,
+      "step": 845
+    },
+    {
+      "epoch": 0.5798090040927695,
+      "grad_norm": 0.9101418256759644,
+      "learning_rate": 7.204460340570658e-05,
+      "loss": 0.01103741154074669,
+      "step": 850
+    },
+    {
+      "epoch": 0.5832196452933152,
+      "grad_norm": 0.22701004147529602,
+      "learning_rate": 7.200458448292972e-05,
+      "loss": 0.06184377670288086,
+      "step": 855
+    },
+    {
+      "epoch": 0.5866302864938608,
+      "grad_norm": 2.3091611862182617,
+      "learning_rate": 7.196430769952126e-05,
+      "loss": 0.0492431253194809,
+      "step": 860
+    },
+    {
+      "epoch": 0.5900409276944065,
+      "grad_norm": 0.6630973815917969,
+      "learning_rate": 7.192377335647876e-05,
+      "loss": 0.027876955270767213,
+      "step": 865
+    },
+    {
+      "epoch": 0.5934515688949522,
+      "grad_norm": 1.7400578260421753,
+      "learning_rate": 7.188298175672464e-05,
+      "loss": 0.023742210865020753,
+      "step": 870
+    },
+    {
+      "epoch": 0.596862210095498,
+      "grad_norm": 0.7121092081069946,
+      "learning_rate": 7.184193320510379e-05,
+      "loss": 0.02125793993473053,
+      "step": 875
+    },
+    {
+      "epoch": 0.6002728512960437,
+      "grad_norm": 0.429611474275589,
+      "learning_rate": 7.180062800838143e-05,
+      "loss": 0.06682519316673279,
+      "step": 880
+    },
+    {
+      "epoch": 0.6036834924965894,
+      "grad_norm": 0.018405891954898834,
+      "learning_rate": 7.17590664752407e-05,
+      "loss": 0.022519922256469725,
+      "step": 885
+    },
+    {
+      "epoch": 0.607094133697135,
+      "grad_norm": 0.9797202348709106,
+      "learning_rate": 7.171724891628046e-05,
+      "loss": 0.10513803958892823,
+      "step": 890
+    },
+    {
+      "epoch": 0.6105047748976807,
+      "grad_norm": 0.11931606382131577,
+      "learning_rate": 7.167517564401282e-05,
+      "loss": 0.055521953105926516,
+      "step": 895
+    },
+    {
+      "epoch": 0.6139154160982264,
+      "grad_norm": 0.04398813843727112,
+      "learning_rate": 7.163284697286097e-05,
+      "loss": 0.018342888355255126,
+      "step": 900
+    },
+    {
+      "epoch": 0.6173260572987722,
+      "grad_norm": 0.11505168676376343,
+      "learning_rate": 7.15902632191567e-05,
+      "loss": 0.026946401596069335,
+      "step": 905
+    },
+    {
+      "epoch": 0.6207366984993179,
+      "grad_norm": 0.3042922019958496,
+      "learning_rate": 7.154742470113816e-05,
+      "loss": 0.02314314842224121,
+      "step": 910
+    },
+    {
+      "epoch": 0.6241473396998636,
+      "grad_norm": 0.09922346472740173,
+      "learning_rate": 7.150433173894733e-05,
+      "loss": 0.06378543972969056,
+      "step": 915
+    },
+    {
+      "epoch": 0.6275579809004093,
+      "grad_norm": 0.6513009667396545,
+      "learning_rate": 7.146098465462776e-05,
+      "loss": 0.036993378400802614,
+      "step": 920
+    },
+    {
+      "epoch": 0.630968622100955,
+      "grad_norm": 1.131602168083191,
+      "learning_rate": 7.14173837721221e-05,
+      "loss": 0.09491010308265686,
+      "step": 925
+    },
+    {
+      "epoch": 0.6343792633015006,
+      "grad_norm": 0.1672857254743576,
+      "learning_rate": 7.137352941726969e-05,
+      "loss": 0.03719751834869385,
+      "step": 930
+    },
+    {
+      "epoch": 0.6377899045020464,
+      "grad_norm": 0.7450887560844421,
+      "learning_rate": 7.132942191780414e-05,
+      "loss": 0.028598809242248537,
+      "step": 935
+    },
+    {
+      "epoch": 0.6412005457025921,
+      "grad_norm": 2.3537657260894775,
+      "learning_rate": 7.128506160335084e-05,
+      "loss": 0.06678735613822936,
+      "step": 940
+    },
+    {
+      "epoch": 0.6446111869031378,
+      "grad_norm": 0.014428210444748402,
+      "learning_rate": 7.124044880542455e-05,
+      "loss": 0.018354634940624236,
+      "step": 945
+    },
+    {
+      "epoch": 0.6480218281036835,
+      "grad_norm": 2.9239041805267334,
+      "learning_rate": 7.119558385742688e-05,
+      "loss": 0.08242651224136352,
+      "step": 950
+    },
+    {
+      "epoch": 0.6514324693042292,
+      "grad_norm": 0.39032718539237976,
+      "learning_rate": 7.115046709464383e-05,
+      "loss": 0.023772728443145753,
+      "step": 955
+    },
+    {
+      "epoch": 0.654843110504775,
+      "grad_norm": 0.3000798523426056,
+      "learning_rate": 7.110509885424326e-05,
+      "loss": 0.03464276790618896,
+      "step": 960
+    },
+    {
+      "epoch": 0.6582537517053206,
+      "grad_norm": 0.3980049192905426,
+      "learning_rate": 7.105947947527238e-05,
+      "loss": 0.08540127277374268,
+      "step": 965
+    },
+    {
+      "epoch": 0.6616643929058663,
+      "grad_norm": 0.9492272734642029,
+      "learning_rate": 7.10136092986552e-05,
+      "loss": 0.02300785481929779,
+      "step": 970
+    },
+    {
+      "epoch": 0.665075034106412,
+      "grad_norm": 1.9585710763931274,
+      "learning_rate": 7.096748866719005e-05,
+      "loss": 0.034704044461250305,
+      "step": 975
+    },
+    {
+      "epoch": 0.6684856753069577,
+      "grad_norm": 1.142238974571228,
+      "learning_rate": 7.092111792554689e-05,
+      "loss": 0.01860647052526474,
+      "step": 980
+    },
+    {
+      "epoch": 0.6718963165075034,
+      "grad_norm": 0.8443534970283508,
+      "learning_rate": 7.087449742026488e-05,
+      "loss": 0.03302992284297943,
+      "step": 985
+    },
+    {
+      "epoch": 0.6753069577080492,
+      "grad_norm": 1.0402863025665283,
+      "learning_rate": 7.082762749974968e-05,
+      "loss": 0.03963000178337097,
+      "step": 990
+    },
+    {
+      "epoch": 0.6787175989085948,
+      "grad_norm": 0.11959892511367798,
+      "learning_rate": 7.078050851427089e-05,
+      "loss": 0.0187692254781723,
+      "step": 995
+    },
+    {
+      "epoch": 0.6821282401091405,
+      "grad_norm": 1.495011329650879,
+      "learning_rate": 7.073314081595945e-05,
+      "loss": 0.050608736276626584,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6855388813096862,
+      "grad_norm": 0.2534504234790802,
+      "learning_rate": 7.068552475880499e-05,
+      "loss": 0.0076520174741745,
+      "step": 1005
+    },
+    {
+      "epoch": 0.6889495225102319,
+      "grad_norm": 0.17387191951274872,
+      "learning_rate": 7.063766069865314e-05,
+      "loss": 0.028283193707466125,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6923601637107776,
+      "grad_norm": 0.07424458116292953,
+      "learning_rate": 7.058954899320297e-05,
+      "loss": 0.03078552782535553,
+      "step": 1015
+    },
+    {
+      "epoch": 0.6957708049113234,
+      "grad_norm": 0.5854848027229309,
+      "learning_rate": 7.05411900020042e-05,
+      "loss": 0.02033105492591858,
+      "step": 1020
+    },
+    {
+      "epoch": 0.699181446111869,
+      "grad_norm": 0.09108871966600418,
+      "learning_rate": 7.049258408645463e-05,
+      "loss": 0.0510578989982605,
+      "step": 1025
+    },
+    {
+      "epoch": 0.7025920873124147,
+      "grad_norm": 0.2851751148700714,
+      "learning_rate": 7.044373160979734e-05,
+      "loss": 0.10413439273834228,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7060027285129604,
+      "grad_norm": 0.8032590746879578,
+      "learning_rate": 7.039463293711804e-05,
+      "loss": 0.05853385329246521,
+      "step": 1035
+    },
+    {
+      "epoch": 0.7094133697135061,
+      "grad_norm": 0.1301775723695755,
+      "learning_rate": 7.03452884353423e-05,
+      "loss": 0.051472657918930055,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7128240109140518,
+      "grad_norm": 0.46156957745552063,
+      "learning_rate": 7.029569847323287e-05,
+      "loss": 0.034115567803382874,
+      "step": 1045
+    },
+    {
+      "epoch": 0.7162346521145976,
+      "grad_norm": 1.081560730934143,
+      "learning_rate": 7.02458634213868e-05,
+      "loss": 0.059652507305145264,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7196452933151433,
+      "grad_norm": 0.721208930015564,
+      "learning_rate": 7.019578365223286e-05,
+      "loss": 0.061070340871810916,
+      "step": 1055
+    },
+    {
+      "epoch": 0.723055934515689,
+      "grad_norm": 0.5738947987556458,
+      "learning_rate": 7.014545954002855e-05,
+      "loss": 0.03556577265262604,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7264665757162346,
+      "grad_norm": 0.15053460001945496,
+      "learning_rate": 7.009489146085744e-05,
+      "loss": 0.03284372091293335,
+      "step": 1065
+    },
+    {
+      "epoch": 0.7298772169167803,
+      "grad_norm": 0.4496553838253021,
+      "learning_rate": 7.004407979262635e-05,
+      "loss": 0.07945018410682678,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7332878581173261,
+      "grad_norm": 1.1821213960647583,
+      "learning_rate": 6.999302491506245e-05,
+      "loss": 0.033741748332977294,
+      "step": 1075
+    },
+    {
+      "epoch": 0.7366984993178718,
+      "grad_norm": 0.2809429168701172,
+      "learning_rate": 6.994172720971047e-05,
+      "loss": 0.023005199432373048,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7401091405184175,
+      "grad_norm": 0.14925819635391235,
+      "learning_rate": 6.989018705992991e-05,
+      "loss": 0.01791207939386368,
+      "step": 1085
+    },
+    {
+      "epoch": 0.7435197817189632,
+      "grad_norm": 1.1947131156921387,
+      "learning_rate": 6.983840485089203e-05,
+      "loss": 0.03395574688911438,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7469304229195088,
+      "grad_norm": 1.336547613143921,
+      "learning_rate": 6.978638096957712e-05,
+      "loss": 0.02712726593017578,
+      "step": 1095
+    },
+    {
+      "epoch": 0.7489768076398363,
+      "eval_loss": 0.05635881423950195,
+      "eval_runtime": 0.8951,
+      "eval_samples_per_second": 83.789,
+      "eval_steps_per_second": 2.234,
+      "step": 1098
+    },
+    {
+      "eval_cer_subset": 0.023981468864967978,
+      "eval_cer_subset_edit_distance": 176,
+      "eval_cer_subset_groups": 75,
+      "eval_cer_subset_items": 75,
+      "eval_cer_subset_ref_chars": 7339,
+      "step": 1098
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 5864,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 366,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.1443907853139968e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc6915d8d9dd9b5c9c17756c87ba7ec8221fd06789232d79225f9518167f0aa1
+size 5841

meta-llama__meta-llama-3.1-8b/sft/run_manifest.json

@@ -6,10 +6,10 @@
   "model_name": "meta-llama/Meta-Llama-3.1-8B",
   "status": "running",
   "best_metric_name": "eval_loss",
-  "best_metric_value": 0.061700768768787384,
-  "best_step": 732,
-  "best_checkpoint_step": 732,
-  "latest_checkpoint_step": 732,
+  "best_metric_value": 0.05635881423950195,
+  "best_step": 1098,
+  "best_checkpoint_step": 1098,
+  "latest_checkpoint_step": 1098,
   "checkpoints": [
     {
       "step": 366,
@@ -22,10 +22,16 @@
       "epoch": 0.49931787175989084,
       "repo_path": "meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-732",
       "timestamp": "2026-04-09T22:14:53.515787+00:00"
+    },
+    {
+      "step": 1098,
+      "epoch": 0.7489768076398363,
+      "repo_path": "meta-llama__meta-llama-3.1-8b/sft/checkpoints/checkpoint-1098",
+      "timestamp": "2026-04-09T22:15:42.601728+00:00"
     }
   ],
   "epoch_artifacts": [],
   "resume_history": [],
   "created_at": "2026-04-09T22:12:40.588264+00:00",
-  "updated_at": "2026-04-09T22:14:53.515801+00:00"
+  "updated_at": "2026-04-09T22:15:42.601744+00:00"
 }