Training in progress, step 700, checkpoint

checkpoint-700/README.md

@@ -0,0 +1,204 @@
+---
+library_name: peft
+base_model: HuggingFaceH4/zephyr-7b-beta
+---
+
+# 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 0.10.0

checkpoint-700/adapter_config.json

@@ -0,0 +1,31 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "HuggingFaceH4/zephyr-7b-beta",
+  "bias": "none",
+  "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": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "q_proj",
+    "o_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-700/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9ff12f3f4d5befc156462ce388fd17f745898fa371492d3ca8fd33ef9dd7c58
+size 54560368

checkpoint-700/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4078e3bf90255f3748cc62a96476657abe2f8a9b597496760dbffd88414e2f3d
+size 109267450

checkpoint-700/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9afbf853352cfbcfea61884ff6a2ddcd2aee1ce8618589cf5b56912c1b160011
+size 14244

checkpoint-700/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdff8de1e52c9d9fd36644eaf5659d424dd94dab3d3f541b3733ebd1f7269f60
+size 1064

checkpoint-700/special_tokens_map.json

@@ -0,0 +1,29 @@
+{
+  "additional_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>"
+  ],
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<unk>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-700/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443

checkpoint-700/tokenizer_config.json

@@ -0,0 +1,50 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>"
+  ],
+  "bos_token": "<s>",
+  "chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n'  + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "max_lenght": 8192,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<unk>",
+  "padding": true,
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "truncation_side": "left",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true
+}

checkpoint-700/trainer_state.json

@@ -0,0 +1,672 @@
+{
+  "best_metric": 0.005181881133466959,
+  "best_model_checkpoint": "./Zephyr/28-03-24-Weni-WeniGPT-QA-Zephyr-7B-4.0.2-KTO_WeniGPT Experiment using KTO trainer with no collator, Zephyr model and random system prompt.-2_max_steps-786_batch_32_2024-03-28_ppid_9/checkpoint-700",
+  "epoch": 5.32825880114177,
+  "eval_steps": 50,
+  "global_step": 700,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.15,
+      "grad_norm": 3.973484516143799,
+      "kl": 0.07447954267263412,
+      "learning_rate": 0.00015833333333333332,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 1.3654,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 20
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 2.1150100231170654,
+      "kl": 0.24954533576965332,
+      "learning_rate": 0.00019606299212598428,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.1911,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 40
+    },
+    {
+      "epoch": 0.38,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -157.132080078125,
+      "eval_logps/rejected": -431.1063232421875,
+      "eval_loss": 0.019795970991253853,
+      "eval_rewards/chosen": 5.025023460388184,
+      "eval_rewards/margins": 28.0367374420166,
+      "eval_rewards/rejected": -23.0117130279541,
+      "eval_runtime": 209.568,
+      "eval_samples_per_second": 2.252,
+      "eval_steps_per_second": 0.563,
+      "step": 50
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.8193832039833069,
+      "kl": 0.0,
+      "learning_rate": 0.00019081364829396326,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.1212,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 60
+    },
+    {
+      "epoch": 0.61,
+      "grad_norm": 1.1438875198364258,
+      "kl": 0.0,
+      "learning_rate": 0.00018556430446194227,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0613,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 80
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 1.5549229383468628,
+      "kl": 0.30086809396743774,
+      "learning_rate": 0.00018031496062992125,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0525,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 100
+    },
+    {
+      "epoch": 0.76,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -151.0825958251953,
+      "eval_logps/rejected": -502.1325378417969,
+      "eval_loss": 0.015516282990574837,
+      "eval_rewards/chosen": 5.629973411560059,
+      "eval_rewards/margins": 35.744300842285156,
+      "eval_rewards/rejected": -30.11433219909668,
+      "eval_runtime": 209.5342,
+      "eval_samples_per_second": 2.253,
+      "eval_steps_per_second": 0.563,
+      "step": 100
+    },
+    {
+      "epoch": 0.91,
+      "grad_norm": 0.16322359442710876,
+      "kl": 0.0,
+      "learning_rate": 0.0001750656167979003,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.037,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 120
+    },
+    {
+      "epoch": 1.07,
+      "grad_norm": 0.8122725486755371,
+      "kl": 0.0,
+      "learning_rate": 0.00016981627296587927,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0684,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 140
+    },
+    {
+      "epoch": 1.14,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -145.42347717285156,
+      "eval_logps/rejected": -530.3358154296875,
+      "eval_loss": 0.010358058847486973,
+      "eval_rewards/chosen": 6.19588565826416,
+      "eval_rewards/margins": 39.130550384521484,
+      "eval_rewards/rejected": -32.934661865234375,
+      "eval_runtime": 209.5283,
+      "eval_samples_per_second": 2.253,
+      "eval_steps_per_second": 0.563,
+      "step": 150
+    },
+    {
+      "epoch": 1.22,
+      "grad_norm": 0.11464398354291916,
+      "kl": 0.0,
+      "learning_rate": 0.00016456692913385828,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0167,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 160
+    },
+    {
+      "epoch": 1.37,
+      "grad_norm": 5.260560989379883,
+      "kl": 0.0,
+      "learning_rate": 0.00015931758530183726,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.131,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 180
+    },
+    {
+      "epoch": 1.52,
+      "grad_norm": 0.15409517288208008,
+      "kl": 0.0,
+      "learning_rate": 0.0001540682414698163,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0253,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 200
+    },
+    {
+      "epoch": 1.52,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -147.8525390625,
+      "eval_logps/rejected": -489.1932678222656,
+      "eval_loss": 0.012590945698320866,
+      "eval_rewards/chosen": 5.952979564666748,
+      "eval_rewards/margins": 34.77338790893555,
+      "eval_rewards/rejected": -28.82040786743164,
+      "eval_runtime": 209.5044,
+      "eval_samples_per_second": 2.253,
+      "eval_steps_per_second": 0.563,
+      "step": 200
+    },
+    {
+      "epoch": 1.67,
+      "grad_norm": 0.025852208957076073,
+      "kl": 0.0,
+      "learning_rate": 0.00014908136482939634,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0101,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 220
+    },
+    {
+      "epoch": 1.83,
+      "grad_norm": 0.013888155110180378,
+      "kl": 0.1387307345867157,
+      "learning_rate": 0.00014383202099737535,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0177,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 240
+    },
+    {
+      "epoch": 1.9,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -142.92942810058594,
+      "eval_logps/rejected": -740.499267578125,
+      "eval_loss": 0.01631937175989151,
+      "eval_rewards/chosen": 6.445290565490723,
+      "eval_rewards/margins": 60.39630126953125,
+      "eval_rewards/rejected": -53.951011657714844,
+      "eval_runtime": 209.4415,
+      "eval_samples_per_second": 2.254,
+      "eval_steps_per_second": 0.563,
+      "step": 250
+    },
+    {
+      "epoch": 1.98,
+      "grad_norm": 0.02793472446501255,
+      "kl": 0.0,
+      "learning_rate": 0.00013858267716535433,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0112,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 260
+    },
+    {
+      "epoch": 2.13,
+      "grad_norm": 0.0313153937458992,
+      "kl": 0.0,
+      "learning_rate": 0.00013333333333333334,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0067,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 280
+    },
+    {
+      "epoch": 2.28,
+      "grad_norm": 0.030554356053471565,
+      "kl": 0.0,
+      "learning_rate": 0.00012808398950131235,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0126,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 300
+    },
+    {
+      "epoch": 2.28,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -145.38067626953125,
+      "eval_logps/rejected": -851.4735717773438,
+      "eval_loss": 0.020638220012187958,
+      "eval_rewards/chosen": 6.200164794921875,
+      "eval_rewards/margins": 71.24861145019531,
+      "eval_rewards/rejected": -65.04844665527344,
+      "eval_runtime": 209.5591,
+      "eval_samples_per_second": 2.252,
+      "eval_steps_per_second": 0.563,
+      "step": 300
+    },
+    {
+      "epoch": 2.44,
+      "grad_norm": 0.01577291637659073,
+      "kl": 0.0,
+      "learning_rate": 0.00012283464566929136,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0134,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 320
+    },
+    {
+      "epoch": 2.59,
+      "grad_norm": 0.017728939652442932,
+      "kl": 0.0,
+      "learning_rate": 0.00011758530183727034,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.007,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 340
+    },
+    {
+      "epoch": 2.66,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -140.51016235351562,
+      "eval_logps/rejected": -616.1578369140625,
+      "eval_loss": 0.006825109478086233,
+      "eval_rewards/chosen": 6.687215328216553,
+      "eval_rewards/margins": 48.20407485961914,
+      "eval_rewards/rejected": -41.51686477661133,
+      "eval_runtime": 209.4969,
+      "eval_samples_per_second": 2.253,
+      "eval_steps_per_second": 0.563,
+      "step": 350
+    },
+    {
+      "epoch": 2.74,
+      "grad_norm": 0.025538522750139236,
+      "kl": 0.2244974821805954,
+      "learning_rate": 0.00011233595800524934,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0034,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 360
+    },
+    {
+      "epoch": 2.89,
+      "grad_norm": 0.009798596613109112,
+      "kl": 0.0,
+      "learning_rate": 0.00010708661417322836,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0081,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 380
+    },
+    {
+      "epoch": 3.04,
+      "grad_norm": 0.021639486774802208,
+      "kl": 0.0,
+      "learning_rate": 0.00010183727034120735,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0085,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 400
+    },
+    {
+      "epoch": 3.04,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -139.7056884765625,
+      "eval_logps/rejected": -727.2117309570312,
+      "eval_loss": 0.007565339561551809,
+      "eval_rewards/chosen": 6.767664432525635,
+      "eval_rewards/margins": 59.389923095703125,
+      "eval_rewards/rejected": -52.62226104736328,
+      "eval_runtime": 209.4194,
+      "eval_samples_per_second": 2.254,
+      "eval_steps_per_second": 0.563,
+      "step": 400
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 0.2425842434167862,
+      "kl": 0.0,
+      "learning_rate": 9.658792650918635e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0029,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 420
+    },
+    {
+      "epoch": 3.35,
+      "grad_norm": 0.01065791118890047,
+      "kl": 0.0,
+      "learning_rate": 9.133858267716536e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0057,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 440
+    },
+    {
+      "epoch": 3.43,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -141.96299743652344,
+      "eval_logps/rejected": -582.8775024414062,
+      "eval_loss": 0.005958274472504854,
+      "eval_rewards/chosen": 6.541934967041016,
+      "eval_rewards/margins": 44.73076629638672,
+      "eval_rewards/rejected": -38.18882751464844,
+      "eval_runtime": 209.4524,
+      "eval_samples_per_second": 2.253,
+      "eval_steps_per_second": 0.563,
+      "step": 450
+    },
+    {
+      "epoch": 3.5,
+      "grad_norm": 0.015213225036859512,
+      "kl": 0.0,
+      "learning_rate": 8.608923884514435e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.008,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 460
+    },
+    {
+      "epoch": 3.65,
+      "grad_norm": 0.016072571277618408,
+      "kl": 0.0,
+      "learning_rate": 8.083989501312336e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0119,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 480
+    },
+    {
+      "epoch": 3.81,
+      "grad_norm": 0.007641986012458801,
+      "kl": 0.16172848641872406,
+      "learning_rate": 7.559055118110236e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0029,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 500
+    },
+    {
+      "epoch": 3.81,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -140.1928253173828,
+      "eval_logps/rejected": -571.4075927734375,
+      "eval_loss": 0.0060875071212649345,
+      "eval_rewards/chosen": 6.718951225280762,
+      "eval_rewards/margins": 43.76079177856445,
+      "eval_rewards/rejected": -37.041839599609375,
+      "eval_runtime": 209.4408,
+      "eval_samples_per_second": 2.254,
+      "eval_steps_per_second": 0.563,
+      "step": 500
+    },
+    {
+      "epoch": 3.96,
+      "grad_norm": 0.009444206021726131,
+      "kl": 0.0,
+      "learning_rate": 7.034120734908137e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0061,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 520
+    },
+    {
+      "epoch": 4.11,
+      "grad_norm": 0.008130188100039959,
+      "kl": 0.0,
+      "learning_rate": 6.509186351706036e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.003,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 540
+    },
+    {
+      "epoch": 4.19,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -138.9207000732422,
+      "eval_logps/rejected": -567.5726928710938,
+      "eval_loss": 0.00618448993191123,
+      "eval_rewards/chosen": 6.84616231918335,
+      "eval_rewards/margins": 43.504512786865234,
+      "eval_rewards/rejected": -36.658355712890625,
+      "eval_runtime": 209.4628,
+      "eval_samples_per_second": 2.253,
+      "eval_steps_per_second": 0.563,
+      "step": 550
+    },
+    {
+      "epoch": 4.26,
+      "grad_norm": 0.007174960803240538,
+      "kl": 0.0,
+      "learning_rate": 5.984251968503938e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0043,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 560
+    },
+    {
+      "epoch": 4.41,
+      "grad_norm": 0.009455788880586624,
+      "kl": 0.0,
+      "learning_rate": 5.4593175853018376e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0034,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 580
+    },
+    {
+      "epoch": 4.57,
+      "grad_norm": 0.010507142171263695,
+      "kl": 0.0,
+      "learning_rate": 4.934383202099738e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0023,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 600
+    },
+    {
+      "epoch": 4.57,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -138.7514190673828,
+      "eval_logps/rejected": -583.14013671875,
+      "eval_loss": 0.005779278464615345,
+      "eval_rewards/chosen": 6.863093376159668,
+      "eval_rewards/margins": 45.078189849853516,
+      "eval_rewards/rejected": -38.2150993347168,
+      "eval_runtime": 209.7359,
+      "eval_samples_per_second": 2.25,
+      "eval_steps_per_second": 0.563,
+      "step": 600
+    },
+    {
+      "epoch": 4.72,
+      "grad_norm": 0.0048731728456914425,
+      "kl": 0.16939109563827515,
+      "learning_rate": 4.409448818897638e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0076,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 620
+    },
+    {
+      "epoch": 4.87,
+      "grad_norm": 0.006191925145685673,
+      "kl": 0.0,
+      "learning_rate": 3.8845144356955383e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0028,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 640
+    },
+    {
+      "epoch": 4.95,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -139.12095642089844,
+      "eval_logps/rejected": -604.9989624023438,
+      "eval_loss": 0.0049984036013484,
+      "eval_rewards/chosen": 6.826137065887451,
+      "eval_rewards/margins": 47.227115631103516,
+      "eval_rewards/rejected": -40.400978088378906,
+      "eval_runtime": 209.7256,
+      "eval_samples_per_second": 2.251,
+      "eval_steps_per_second": 0.563,
+      "step": 650
+    },
+    {
+      "epoch": 5.02,
+      "grad_norm": 0.006048476789146662,
+      "kl": 0.0,
+      "learning_rate": 3.3595800524934386e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0099,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 660
+    },
+    {
+      "epoch": 5.18,
+      "grad_norm": 0.0051725320518016815,
+      "kl": 0.0,
+      "learning_rate": 2.8346456692913388e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0033,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 680
+    },
+    {
+      "epoch": 5.33,
+      "grad_norm": 0.009665340185165405,
+      "kl": 0.0,
+      "learning_rate": 2.309711286089239e-05,
+      "logps/chosen": NaN,
+      "logps/rejected": NaN,
+      "loss": 0.0029,
+      "rewards/chosen": NaN,
+      "rewards/margins": NaN,
+      "rewards/rejected": NaN,
+      "step": 700
+    },
+    {
+      "epoch": 5.33,
+      "eval_kl": 0.0,
+      "eval_logps/chosen": -138.82569885253906,
+      "eval_logps/rejected": -603.3299560546875,
+      "eval_loss": 0.005181881133466959,
+      "eval_rewards/chosen": 6.855663776397705,
+      "eval_rewards/margins": 47.089744567871094,
+      "eval_rewards/rejected": -40.234073638916016,
+      "eval_runtime": 209.7627,
+      "eval_samples_per_second": 2.25,
+      "eval_steps_per_second": 0.563,
+      "step": 700
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 786,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 100,
+  "total_flos": 0.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-700/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72ac543b1af4f9675ddc3d87c79e0f49fbd145783e3bc0097b4be05d98c85865
+size 5752