Training in progress, step 2000, checkpoint

checkpoint-2000/README.md

@@ -0,0 +1,209 @@
+---
+base_model: google/medgemma-1.5-4b-it
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:google/medgemma-1.5-4b-it
+- lora
+- sft
+- transformers
+- trl
+---
+
+# 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 0.19.1

checkpoint-2000/adapter_config.json

@@ -0,0 +1,48 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/medgemma-1.5-4b-it",
+  "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": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "k_proj",
+    "o_proj",
+    "q_proj",
+    "v_proj",
+    "gate_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoint-2000/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9a3140b6600448f8164add44d0a99083e4531b78d43f88c938070200516b3a2
+size 131250184

checkpoint-2000/chat_template.jinja

@@ -0,0 +1,47 @@
+{{ bos_token }}
+{%- if messages[0]['role'] == 'system' -%}
+    {%- if messages[0]['content'] is string -%}
+        {%- set first_user_prefix = messages[0]['content'] + '
+
+' -%}
+    {%- else -%}
+        {%- set first_user_prefix = messages[0]['content'][0]['text'] + '
+
+' -%}
+    {%- endif -%}
+    {%- set loop_messages = messages[1:] -%}
+{%- else -%}
+    {%- set first_user_prefix = "" -%}
+    {%- set loop_messages = messages -%}
+{%- endif -%}
+{%- for message in loop_messages -%}
+    {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
+        {{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
+    {%- endif -%}
+    {%- if (message['role'] == 'assistant') -%}
+        {%- set role = "model" -%}
+    {%- else -%}
+        {%- set role = message['role'] -%}
+    {%- endif -%}
+    {{ '<start_of_turn>' + role + '
+' + (first_user_prefix if loop.first else "") }}
+    {%- if message['content'] is string -%}
+        {{ message['content'] | trim }}
+    {%- elif message['content'] is iterable -%}
+        {%- for item in message['content'] -%}
+            {%- if item['type'] == 'image' -%}
+                {{ '<start_of_image>' }}
+            {%- elif item['type'] == 'text' -%}
+                {{ item['text'] | trim }}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- else -%}
+        {{ raise_exception("Invalid content type") }}
+    {%- endif -%}
+    {{ '<end_of_turn>
+' }}
+{%- endfor -%}
+{%- if add_generation_prompt -%}
+    {{'<start_of_turn>model
+'}}
+{%- endif -%}

checkpoint-2000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30e9e59ce5d9e927a3cfeebc18c97abd47ed60a7fb64069b1018b77e258bcd59
+size 238828343

checkpoint-2000/processor_config.json

@@ -0,0 +1,28 @@
+{
+  "image_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "Gemma3ImageProcessor",
+    "image_seq_length": 256,
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "resample": 2,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "height": 896,
+      "width": 896
+    }
+  },
+  "image_seq_length": 256,
+  "processor_class": "Gemma3Processor"
+}

checkpoint-2000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38b1d407e20aa6c962f308a3a7464d5113fcb298aae739cfdbae4a929163cb14
+size 14645

checkpoint-2000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7329f48ebc4aca18f34446dab7a52e1edc3bfa1a7e9e26840152d707cc017892
+size 1465

checkpoint-2000/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6c776bcef3fcea90b4244ef4dafa43ec828ea9f23a286a5be6bd3c209efe810
+size 33384721

checkpoint-2000/tokenizer_config.json

@@ -0,0 +1,26 @@
+{
+  "backend": "pil",
+  "boi_token": "<start_of_image>",
+  "bos_token": "<bos>",
+  "clean_up_tokenization_spaces": false,
+  "eoi_token": "<end_of_image>",
+  "eos_token": "<eos>",
+  "image_token": "<image_soft_token>",
+  "is_local": false,
+  "local_files_only": false,
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "model_specific_special_tokens": {
+    "boi_token": "<start_of_image>",
+    "eoi_token": "<end_of_image>",
+    "image_token": "<image_soft_token>"
+  },
+  "pad_token": "<pad>",
+  "padding_side": "left",
+  "processor_class": "Gemma3Processor",
+  "sp_model_kwargs": null,
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

checkpoint-2000/trainer_state.json

@@ -0,0 +1,544 @@
+{
+  "best_global_step": 2000,
+  "best_metric": 0.14112941920757294,
+  "best_model_checkpoint": "data/full_sft_run/sft_out/checkpoint-2000",
+  "epoch": 0.5845816587504568,
+  "eval_steps": 200,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 0.20752533283084632,
+      "epoch": 0.014614541468761417,
+      "grad_norm": 0.7449454069137573,
+      "learning_rate": 3.181818181818182e-05,
+      "loss": 1.3096963500976562,
+      "mean_token_accuracy": 0.8356087231636047,
+      "num_tokens": 218234.0,
+      "step": 50
+    },
+    {
+      "entropy": 0.2586296512931585,
+      "epoch": 0.029229082937522834,
+      "grad_norm": 0.6199880838394165,
+      "learning_rate": 6.428571428571429e-05,
+      "loss": 0.26672126770019533,
+      "mean_token_accuracy": 0.9052031901478768,
+      "num_tokens": 434925.0,
+      "step": 100
+    },
+    {
+      "entropy": 0.2025177276879549,
+      "epoch": 0.043843624406284254,
+      "grad_norm": 0.5352309346199036,
+      "learning_rate": 9.675324675324677e-05,
+      "loss": 0.20794065475463866,
+      "mean_token_accuracy": 0.9208106356859207,
+      "num_tokens": 652823.0,
+      "step": 150
+    },
+    {
+      "entropy": 0.18583494622260333,
+      "epoch": 0.05845816587504567,
+      "grad_norm": 0.4790455996990204,
+      "learning_rate": 0.00012922077922077922,
+      "loss": 0.19879446029663086,
+      "mean_token_accuracy": 0.9228935891389847,
+      "num_tokens": 870027.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.05845816587504567,
+      "eval_entropy": 0.20352977880481946,
+      "eval_loss": 0.1885823905467987,
+      "eval_mean_token_accuracy": 0.9265667232416444,
+      "eval_num_tokens": 870027.0,
+      "eval_runtime": 51.9934,
+      "eval_samples_per_second": 22.657,
+      "eval_steps_per_second": 5.674,
+      "step": 200
+    },
+    {
+      "entropy": 0.18395513169467448,
+      "epoch": 0.0730727073438071,
+      "grad_norm": 0.5151433348655701,
+      "learning_rate": 0.0001616883116883117,
+      "loss": 0.1874466896057129,
+      "mean_token_accuracy": 0.9266703170537949,
+      "num_tokens": 1086768.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.18530432548373937,
+      "epoch": 0.08768724881256851,
+      "grad_norm": 0.360085666179657,
+      "learning_rate": 0.00019415584415584416,
+      "loss": 0.18927602767944335,
+      "mean_token_accuracy": 0.9263237661123276,
+      "num_tokens": 1306053.0,
+      "step": 300
+    },
+    {
+      "entropy": 0.18520764522254468,
+      "epoch": 0.10230179028132992,
+      "grad_norm": 0.32682710886001587,
+      "learning_rate": 0.00019917654147419162,
+      "loss": 0.19010570526123047,
+      "mean_token_accuracy": 0.9269409370422363,
+      "num_tokens": 1524650.0,
+      "step": 350
+    },
+    {
+      "entropy": 0.17121127102524042,
+      "epoch": 0.11691633175009133,
+      "grad_norm": 0.33935022354125977,
+      "learning_rate": 0.00019817232375979112,
+      "loss": 0.17950326919555665,
+      "mean_token_accuracy": 0.928639045059681,
+      "num_tokens": 1740391.0,
+      "step": 400
+    },
+    {
+      "epoch": 0.11691633175009133,
+      "eval_entropy": 0.15643637586953277,
+      "eval_loss": 0.17647278308868408,
+      "eval_mean_token_accuracy": 0.9313437746743024,
+      "eval_num_tokens": 1740391.0,
+      "eval_runtime": 51.4028,
+      "eval_samples_per_second": 22.917,
+      "eval_steps_per_second": 5.739,
+      "step": 400
+    },
+    {
+      "entropy": 0.17545538306236266,
+      "epoch": 0.13153087321885276,
+      "grad_norm": 0.35228046774864197,
+      "learning_rate": 0.00019716810604539066,
+      "loss": 0.18172544479370117,
+      "mean_token_accuracy": 0.9291265362501144,
+      "num_tokens": 1957727.0,
+      "step": 450
+    },
+    {
+      "entropy": 0.1679905788227916,
+      "epoch": 0.1461454146876142,
+      "grad_norm": 0.36982566118240356,
+      "learning_rate": 0.0001961638883309902,
+      "loss": 0.17352703094482422,
+      "mean_token_accuracy": 0.9319902381300926,
+      "num_tokens": 2174741.0,
+      "step": 500
+    },
+    {
+      "entropy": 0.16595628045499325,
+      "epoch": 0.1607599561563756,
+      "grad_norm": 0.2658090889453888,
+      "learning_rate": 0.0001951596706165897,
+      "loss": 0.1694266128540039,
+      "mean_token_accuracy": 0.9349100396037102,
+      "num_tokens": 2392020.0,
+      "step": 550
+    },
+    {
+      "entropy": 0.16491649385541676,
+      "epoch": 0.17537449762513702,
+      "grad_norm": 0.28709471225738525,
+      "learning_rate": 0.0001941554529021892,
+      "loss": 0.16749130249023436,
+      "mean_token_accuracy": 0.9363178312778473,
+      "num_tokens": 2609502.0,
+      "step": 600
+    },
+    {
+      "epoch": 0.17537449762513702,
+      "eval_entropy": 0.15555994159589379,
+      "eval_loss": 0.16351227462291718,
+      "eval_mean_token_accuracy": 0.935047954623982,
+      "eval_num_tokens": 2609502.0,
+      "eval_runtime": 51.7247,
+      "eval_samples_per_second": 22.774,
+      "eval_steps_per_second": 5.703,
+      "step": 600
+    },
+    {
+      "entropy": 0.16599890314042567,
+      "epoch": 0.18998903909389842,
+      "grad_norm": 0.3486684560775757,
+      "learning_rate": 0.00019315123518778873,
+      "loss": 0.16908374786376953,
+      "mean_token_accuracy": 0.9338638958334923,
+      "num_tokens": 2826468.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.15440633669495582,
+      "epoch": 0.20460358056265984,
+      "grad_norm": 0.3269498944282532,
+      "learning_rate": 0.00019214701747338823,
+      "loss": 0.1564903450012207,
+      "mean_token_accuracy": 0.9378401437401771,
+      "num_tokens": 3044037.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.15702761143445967,
+      "epoch": 0.21921812203142127,
+      "grad_norm": 0.2996591329574585,
+      "learning_rate": 0.00019114279975898776,
+      "loss": 0.16098966598510742,
+      "mean_token_accuracy": 0.937047846019268,
+      "num_tokens": 3260671.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.16964259944856167,
+      "epoch": 0.23383266350018267,
+      "grad_norm": 0.27347952127456665,
+      "learning_rate": 0.00019013858204458727,
+      "loss": 0.17197874069213867,
+      "mean_token_accuracy": 0.9319348493218422,
+      "num_tokens": 3479952.0,
+      "step": 800
+    },
+    {
+      "epoch": 0.23383266350018267,
+      "eval_entropy": 0.15915015558317555,
+      "eval_loss": 0.15902064740657806,
+      "eval_mean_token_accuracy": 0.9373329009039927,
+      "eval_num_tokens": 3479952.0,
+      "eval_runtime": 52.0586,
+      "eval_samples_per_second": 22.628,
+      "eval_steps_per_second": 5.667,
+      "step": 800
+    },
+    {
+      "entropy": 0.16319214086979628,
+      "epoch": 0.2484472049689441,
+      "grad_norm": 0.2520512342453003,
+      "learning_rate": 0.0001891343643301868,
+      "loss": 0.16472122192382813,
+      "mean_token_accuracy": 0.9351362636685372,
+      "num_tokens": 3695714.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.157652658931911,
+      "epoch": 0.2630617464377055,
+      "grad_norm": 0.2863970398902893,
+      "learning_rate": 0.00018813014661578633,
+      "loss": 0.15824624061584472,
+      "mean_token_accuracy": 0.9369783601164818,
+      "num_tokens": 3913429.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.15214558444917203,
+      "epoch": 0.2776762879064669,
+      "grad_norm": 0.290787935256958,
+      "learning_rate": 0.00018712592890138583,
+      "loss": 0.15539746284484862,
+      "mean_token_accuracy": 0.9397245013713836,
+      "num_tokens": 4131640.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.15566726494580507,
+      "epoch": 0.2922908293752284,
+      "grad_norm": 0.3074010908603668,
+      "learning_rate": 0.00018612171118698534,
+      "loss": 0.15537342071533203,
+      "mean_token_accuracy": 0.9387755367159843,
+      "num_tokens": 4348832.0,
+      "step": 1000
+    },
+    {
+      "epoch": 0.2922908293752284,
+      "eval_entropy": 0.1574143134436365,
+      "eval_loss": 0.15738964080810547,
+      "eval_mean_token_accuracy": 0.9374752864999286,
+      "eval_num_tokens": 4348832.0,
+      "eval_runtime": 51.5665,
+      "eval_samples_per_second": 22.844,
+      "eval_steps_per_second": 5.721,
+      "step": 1000
+    },
+    {
+      "entropy": 0.1583068972080946,
+      "epoch": 0.3069053708439898,
+      "grad_norm": 0.3427266776561737,
+      "learning_rate": 0.00018511749347258487,
+      "loss": 0.15989484786987304,
+      "mean_token_accuracy": 0.9373269161581993,
+      "num_tokens": 4565702.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.1511041523516178,
+      "epoch": 0.3215199123127512,
+      "grad_norm": 0.32759514451026917,
+      "learning_rate": 0.00018411327575818437,
+      "loss": 0.15331138610839845,
+      "mean_token_accuracy": 0.9397411552071572,
+      "num_tokens": 4784167.0,
+      "step": 1100
+    },
+    {
+      "entropy": 0.14996831517666578,
+      "epoch": 0.33613445378151263,
+      "grad_norm": 0.30234530568122864,
+      "learning_rate": 0.0001831090580437839,
+      "loss": 0.1544932556152344,
+      "mean_token_accuracy": 0.9399288591742515,
+      "num_tokens": 4999793.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.15582640528678893,
+      "epoch": 0.35074899525027403,
+      "grad_norm": 0.24690158665180206,
+      "learning_rate": 0.0001821048403293834,
+      "loss": 0.15442598342895508,
+      "mean_token_accuracy": 0.9398289120197296,
+      "num_tokens": 5217954.0,
+      "step": 1200
+    },
+    {
+      "epoch": 0.35074899525027403,
+      "eval_entropy": 0.1420629285275936,
+      "eval_loss": 0.14971356093883514,
+      "eval_mean_token_accuracy": 0.9414400791717787,
+      "eval_num_tokens": 5217954.0,
+      "eval_runtime": 51.4751,
+      "eval_samples_per_second": 22.885,
+      "eval_steps_per_second": 5.731,
+      "step": 1200
+    },
+    {
+      "entropy": 0.1543054273724556,
+      "epoch": 0.36536353671903543,
+      "grad_norm": 0.29708531498908997,
+      "learning_rate": 0.00018110062261498294,
+      "loss": 0.1537049388885498,
+      "mean_token_accuracy": 0.9416613772511482,
+      "num_tokens": 5434380.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.14713672876358033,
+      "epoch": 0.37997807818779683,
+      "grad_norm": 0.2888696789741516,
+      "learning_rate": 0.00018009640490058247,
+      "loss": 0.15287999153137208,
+      "mean_token_accuracy": 0.9416247588396073,
+      "num_tokens": 5651784.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.14809815444052218,
+      "epoch": 0.3945926196565583,
+      "grad_norm": 0.25226420164108276,
+      "learning_rate": 0.00017909218718618197,
+      "loss": 0.15003498077392577,
+      "mean_token_accuracy": 0.9424466270208359,
+      "num_tokens": 5870443.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.15500445548444985,
+      "epoch": 0.4092071611253197,
+      "grad_norm": 0.2410505712032318,
+      "learning_rate": 0.0001780879694717815,
+      "loss": 0.15515862464904784,
+      "mean_token_accuracy": 0.9403725653886795,
+      "num_tokens": 6086417.0,
+      "step": 1400
+    },
+    {
+      "epoch": 0.4092071611253197,
+      "eval_entropy": 0.13966038472066492,
+      "eval_loss": 0.1482466757297516,
+      "eval_mean_token_accuracy": 0.9430661270173929,
+      "eval_num_tokens": 6086417.0,
+      "eval_runtime": 51.3223,
+      "eval_samples_per_second": 22.953,
+      "eval_steps_per_second": 5.748,
+      "step": 1400
+    },
+    {
+      "entropy": 0.14597053304314614,
+      "epoch": 0.4238217025940811,
+      "grad_norm": 0.2526480257511139,
+      "learning_rate": 0.000177083751757381,
+      "loss": 0.1495055389404297,
+      "mean_token_accuracy": 0.9429304957389831,
+      "num_tokens": 6304323.0,
+      "step": 1450
+    },
+    {
+      "entropy": 0.1473805220425129,
+      "epoch": 0.43843624406284254,
+      "grad_norm": 0.27721503376960754,
+      "learning_rate": 0.0001760795340429805,
+      "loss": 0.14802109718322753,
+      "mean_token_accuracy": 0.9423278120160102,
+      "num_tokens": 6521073.0,
+      "step": 1500
+    },
+    {
+      "entropy": 0.15090686846524476,
+      "epoch": 0.45305078553160394,
+      "grad_norm": 0.3168330192565918,
+      "learning_rate": 0.00017507531632858004,
+      "loss": 0.1520420265197754,
+      "mean_token_accuracy": 0.9409908491373062,
+      "num_tokens": 6738178.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.14193191647529602,
+      "epoch": 0.46766532700036534,
+      "grad_norm": 0.28218454122543335,
+      "learning_rate": 0.00017407109861417957,
+      "loss": 0.1442671012878418,
+      "mean_token_accuracy": 0.944555883705616,
+      "num_tokens": 6955413.0,
+      "step": 1600
+    },
+    {
+      "epoch": 0.46766532700036534,
+      "eval_entropy": 0.14087487072762797,
+      "eval_loss": 0.14355963468551636,
+      "eval_mean_token_accuracy": 0.943681773694895,
+      "eval_num_tokens": 6955413.0,
+      "eval_runtime": 52.1208,
+      "eval_samples_per_second": 22.601,
+      "eval_steps_per_second": 5.66,
+      "step": 1600
+    },
+    {
+      "entropy": 0.1416533422470093,
+      "epoch": 0.4822798684691268,
+      "grad_norm": 0.28569167852401733,
+      "learning_rate": 0.00017306688089977908,
+      "loss": 0.146381254196167,
+      "mean_token_accuracy": 0.9433900326490402,
+      "num_tokens": 7171960.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.14460256807506083,
+      "epoch": 0.4968944099378882,
+      "grad_norm": 0.298966646194458,
+      "learning_rate": 0.0001720626631853786,
+      "loss": 0.14527856826782226,
+      "mean_token_accuracy": 0.9446076577901841,
+      "num_tokens": 7387948.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.14507481291890145,
+      "epoch": 0.5115089514066496,
+      "grad_norm": 0.32264840602874756,
+      "learning_rate": 0.0001710584454709781,
+      "loss": 0.14816983222961425,
+      "mean_token_accuracy": 0.9431060332059861,
+      "num_tokens": 7605190.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.1457299491018057,
+      "epoch": 0.526123492875411,
+      "grad_norm": 0.23384633660316467,
+      "learning_rate": 0.00017005422775657764,
+      "loss": 0.14449834823608398,
+      "mean_token_accuracy": 0.9444263905286789,
+      "num_tokens": 7822465.0,
+      "step": 1800
+    },
+    {
+      "epoch": 0.526123492875411,
+      "eval_entropy": 0.13727282429650678,
+      "eval_loss": 0.14151474833488464,
+      "eval_mean_token_accuracy": 0.9445956945419312,
+      "eval_num_tokens": 7822465.0,
+      "eval_runtime": 51.7253,
+      "eval_samples_per_second": 22.774,
+      "eval_steps_per_second": 5.703,
+      "step": 1800
+    },
+    {
+      "entropy": 0.13303045216947795,
+      "epoch": 0.5407380343441724,
+      "grad_norm": 0.24926069378852844,
+      "learning_rate": 0.00016905001004217715,
+      "loss": 0.1377289867401123,
+      "mean_token_accuracy": 0.9460858461260796,
+      "num_tokens": 8039013.0,
+      "step": 1850
+    },
+    {
+      "entropy": 0.14333279909566043,
+      "epoch": 0.5553525758129338,
+      "grad_norm": 0.23103737831115723,
+      "learning_rate": 0.00016804579232777665,
+      "loss": 0.14504934310913087,
+      "mean_token_accuracy": 0.9442342269420624,
+      "num_tokens": 8255914.0,
+      "step": 1900
+    },
+    {
+      "entropy": 0.14443909596651794,
+      "epoch": 0.5699671172816952,
+      "grad_norm": 0.3567982017993927,
+      "learning_rate": 0.00016704157461337618,
+      "loss": 0.14560423851013182,
+      "mean_token_accuracy": 0.9445012074708938,
+      "num_tokens": 8473895.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.1381967855617404,
+      "epoch": 0.5845816587504568,
+      "grad_norm": 0.32823431491851807,
+      "learning_rate": 0.00016603735689897571,
+      "loss": 0.1393825626373291,
+      "mean_token_accuracy": 0.9458042901754379,
+      "num_tokens": 8690259.0,
+      "step": 2000
+    },
+    {
+      "epoch": 0.5845816587504568,
+      "eval_entropy": 0.12425902821755005,
+      "eval_loss": 0.14112941920757294,
+      "eval_mean_token_accuracy": 0.9466315289675179,
+      "eval_num_tokens": 8690259.0,
+      "eval_runtime": 51.7824,
+      "eval_samples_per_second": 22.749,
+      "eval_steps_per_second": 5.697,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 10266,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.153002426376256e+17,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a62e955b4a049523c45a1102ae6e2f9b52bf056eeb0fc0b8d413459417f3ac39
+size 5841