training logs and plots

training/orgos-training/orgos_lora_adapter/README.md

@@ -0,0 +1,210 @@
+---
+base_model: unsloth/Qwen2.5-3B-Instruct-bnb-4bit
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:unsloth/Qwen2.5-3B-Instruct-bnb-4bit
+- grpo
+- lora
+- 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 0.19.1

training/orgos-training/orgos_lora_adapter/adapter_config.json

@@ -0,0 +1,52 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen2ForCausalLM",
+    "parent_library": "transformers.models.qwen2.modeling_qwen2",
+    "unsloth_fixed": true
+  },
+  "base_model_name_or_path": "unsloth/Qwen2.5-3B-Instruct-bnb-4bit",
+  "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,
+  "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",
+    "q_proj",
+    "gate_proj",
+    "k_proj",
+    "v_proj",
+    "up_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

training/orgos-training/orgos_lora_adapter/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

training/orgos-training/orgos_lora_adapter/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd5948af71b4f56cf697f7580814c7ce8b80595ef985544efcacf716126a2e31
+size 11422356

training/orgos-training/orgos_lora_adapter/tokenizer_config.json

@@ -0,0 +1,201 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "is_local": false,
+  "model_max_length": 32768,
+  "pad_token": "<|PAD_TOKEN|>",
+  "padding_side": "left",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<|PAD_TOKEN|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    }
+  }
+}

training/orgos-training/training_log.txt

@@ -0,0 +1,134 @@
+[ENV] status=healthy version=2.0.0
+[TRAIN_CONFIG] model=unsloth/Qwen2.5-3B-Instruct-bnb-4bit lora_r=16 max_seq_len=4096 trainable_params=29,933,568 quantization=4bit
+[EVAL_START] phase=baseline
+[EVAL] phase=baseline workflow=A episode=1 score=0.0010
+[EVAL] phase=baseline workflow=A episode=2 score=0.2500
+[EVAL] phase=baseline workflow=A episode=3 score=0.2375
+[EVAL] phase=baseline workflow=A episode=4 score=0.0500
+[EVAL] phase=baseline workflow=A episode=5 score=0.2470
+[EVAL_WORKFLOW] phase=baseline workflow=A mean=0.1571 min=0.0010 max=0.2500
+[EVAL] phase=baseline workflow=B episode=1 score=0.2500
+[EVAL] phase=baseline workflow=B episode=2 score=0.2500
+[EVAL] phase=baseline workflow=B episode=3 score=0.2500
+[EVAL] phase=baseline workflow=B episode=4 score=0.2500
+[EVAL] phase=baseline workflow=B episode=5 score=0.2500
+[EVAL_WORKFLOW] phase=baseline workflow=B mean=0.2500 min=0.2500 max=0.2500
+[EVAL] phase=baseline workflow=C episode=1 score=0.2230
+[EVAL] phase=baseline workflow=C episode=2 score=0.0595
+[EVAL] phase=baseline workflow=C episode=3 score=0.2230
+[EVAL] phase=baseline workflow=C episode=4 score=0.0595
+[EVAL] phase=baseline workflow=C episode=5 score=0.0010
+[EVAL_WORKFLOW] phase=baseline workflow=C mean=0.1132 min=0.0010 max=0.2230
+[EVAL_END] phase=baseline overall_mean=0.1734
+[TRAIN_CONFIG] algorithm=GRPO prompts=60 workflows=A,B,C prompts_per_workflow=20
+[PROMPT_DEBUG] first_prompt_tokens=1303
+[REWARD_FN_CHECK] valid_action=0.0490 invalid_action=-0.1000
+[TRAIN_CONFIG] model=unsloth/Qwen2.5-3B-Instruct-bnb-4bit lora_r=16 max_seq_len=4096 trainable_params=29,933,568 quantization=4bit
+[TRAIN_START] max_steps=80 G=4 lr=5e-06
+[TRAIN_STEP] step=1 reward=-0.0325 loss=-0.0973 kl=0.0000
+[TRAIN_STEP] step=2 reward=0.0615 loss=0.0000 kl=0.0000
+[TRAIN_STEP] step=3 reward=0.0436 loss=0.2717 kl=0.0000
+[TRAIN_STEP] step=4 reward=0.0195 loss=0.0902 kl=0.0000
+[TRAIN_STEP] step=5 reward=0.0258 loss=-0.0360 kl=0.0000
+[TRAIN_STEP] step=6 reward=-0.1121 loss=0.2567 kl=0.0000
+[TRAIN_STEP] step=7 reward=0.0033 loss=-0.0239 kl=0.0000
+[TRAIN_STEP] step=8 reward=-0.1750 loss=0.0451 kl=0.0000
+[TRAIN_STEP] step=9 reward=-0.0971 loss=0.2075 kl=0.0000
+[TRAIN_STEP] step=10 reward=0.0436 loss=-0.0145 kl=0.0001
+[TRAIN_STEP] step=11 reward=0.0033 loss=0.2299 kl=0.0001
+[TRAIN_STEP] step=12 reward=-0.1121 loss=0.2794 kl=0.0001
+[TRAIN_STEP] step=13 reward=-0.1150 loss=0.0157 kl=0.0001
+[TRAIN_STEP] step=14 reward=-0.0030 loss=0.1447 kl=0.0001
+[TRAIN_STEP] step=15 reward=0.0258 loss=0.0529 kl=0.0008
+[TRAIN_STEP] step=16 reward=-0.0146 loss=0.0710 kl=0.0002
+[TRAIN_STEP] step=17 reward=-0.0178 loss=0.0715 kl=0.0005
+[TRAIN_STEP] step=18 reward=0.0195 loss=0.0405 kl=0.0003
+[TRAIN_STEP] step=19 reward=0.0047 loss=0.0000 kl=0.0006
+[TRAIN_STEP] step=20 reward=0.0195 loss=0.1379 kl=0.0015
+[TRAIN_STEP] step=21 reward=0.0033 loss=0.2259 kl=0.0015
+[TRAIN_STEP] step=22 reward=-0.0178 loss=-0.0875 kl=0.0016
+[TRAIN_STEP] step=23 reward=-0.0746 loss=-0.0161 kl=0.0005
+[TRAIN_STEP] step=24 reward=0.0615 loss=0.0000 kl=0.0023
+[TRAIN_STEP] step=25 reward=-0.0700 loss=0.1665 kl=0.0016
+[TRAIN_STEP] step=26 reward=0.0615 loss=0.0000 kl=0.0021
+[TRAIN_STEP] step=27 reward=0.0615 loss=0.0000 kl=0.0017
+[TRAIN_STEP] step=28 reward=0.0195 loss=0.0429 kl=0.0033
+[TRAIN_STEP] step=29 reward=0.0211 loss=0.2210 kl=0.0063
+[TRAIN_STEP] step=30 reward=0.0436 loss=-0.0804 kl=0.0046
+[TRAIN_STEP] step=31 reward=0.0342 loss=0.0000 kl=0.0019
+[TRAIN_STEP] step=32 reward=-0.0521 loss=0.1682 kl=0.0031
+[TRAIN_STEP] step=33 reward=0.0342 loss=0.1024 kl=0.0052
+[TRAIN_STEP] step=34 reward=0.0342 loss=0.0000 kl=0.0030
+[TRAIN_STEP] step=35 reward=0.0436 loss=0.0979 kl=0.0030
+[TRAIN_STEP] step=36 reward=0.0258 loss=0.3306 kl=0.0018
+[TRAIN_STEP] step=37 reward=0.0490 loss=0.0000 kl=0.0019
+[TRAIN_STEP] step=38 reward=0.0615 loss=0.0000 kl=0.0037
+[TRAIN_STEP] step=39 reward=0.0615 loss=0.0000 kl=0.0024
+[TRAIN_STEP] step=40 reward=0.0436 loss=0.0039 kl=0.0163
+[TRAIN_STEP] step=41 reward=0.0342 loss=0.0000 kl=0.0141
+[TRAIN_STEP] step=42 reward=0.0436 loss=0.1721 kl=0.0072
+[TRAIN_STEP] step=43 reward=0.0615 loss=0.0000 kl=0.0037
+[TRAIN_STEP] step=44 reward=0.0615 loss=0.0000 kl=0.0122
+[TRAIN_STEP] step=45 reward=0.0342 loss=0.1661 kl=0.0238
+[TRAIN_STEP] step=46 reward=-0.1900 loss=0.1922 kl=0.0140
+[TRAIN_STEP] step=47 reward=0.0436 loss=0.0049 kl=0.0251
+[TRAIN_STEP] step=48 reward=0.0342 loss=-0.0395 kl=0.0059
+[TRAIN_STEP] step=49 reward=0.0436 loss=-0.0437 kl=0.0128
+[TRAIN_STEP] step=50 reward=-0.1900 loss=0.1363 kl=0.0156
+[TRAIN_STEP] step=51 reward=0.0342 loss=0.0000 kl=0.0210
+[TRAIN_STEP] step=52 reward=-0.0178 loss=0.0326 kl=0.0135
+[TRAIN_STEP] step=53 reward=0.0342 loss=0.0000 kl=0.0298
+[TRAIN_STEP] step=54 reward=-0.0146 loss=0.4537 kl=0.0119
+[TRAIN_STEP] step=55 reward=-0.0342 loss=0.2914 kl=0.0059
+[TRAIN_STEP] step=56 reward=-0.0164 loss=0.1776 kl=0.0142
+[TRAIN_STEP] step=57 reward=0.0342 loss=-0.0275 kl=0.0142
+[TRAIN_STEP] step=58 reward=0.0615 loss=0.0000 kl=0.0224
+[TRAIN_STEP] step=59 reward=0.0211 loss=0.1406 kl=0.0063
+[TRAIN_STEP] step=60 reward=0.0615 loss=0.0000 kl=0.0121
+[TRAIN_STEP] step=61 reward=-0.0192 loss=0.3104 kl=0.0110
+[TRAIN_STEP] step=62 reward=0.0490 loss=0.0000 kl=0.0122
+[TRAIN_STEP] step=63 reward=0.0615 loss=0.0000 kl=0.0005
+[TRAIN_STEP] step=64 reward=-0.1121 loss=0.2547 kl=0.0144
+[TRAIN_STEP] step=65 reward=0.0195 loss=0.0000 kl=0.0133
+[TRAIN_STEP] step=66 reward=-0.0402 loss=0.0749 kl=0.0122
+[TRAIN_STEP] step=67 reward=-0.0371 loss=0.5726 kl=0.0084
+[TRAIN_STEP] step=68 reward=0.0033 loss=0.2116 kl=0.0067
+[TRAIN_STEP] step=69 reward=0.0615 loss=0.0000 kl=0.0161
+[TRAIN_STEP] step=70 reward=0.0615 loss=0.0000 kl=0.0180
+[TRAIN_STEP] step=71 reward=0.0490 loss=0.0000 kl=0.0041
+[TRAIN_STEP] step=72 reward=0.0615 loss=0.0000 kl=0.0414
+[TRAIN_STEP] step=73 reward=0.0342 loss=-0.0453 kl=0.0413
+[TRAIN_STEP] step=74 reward=0.0436 loss=0.1321 kl=0.0039
+[TRAIN_STEP] step=75 reward=0.0342 loss=0.0000 kl=0.0045
+[TRAIN_STEP] step=76 reward=0.0079 loss=-0.0503 kl=0.0047
+[TRAIN_STEP] step=77 reward=0.0615 loss=0.0000 kl=0.0243
+[TRAIN_STEP] step=78 reward=0.0490 loss=0.0000 kl=0.0026
+[TRAIN_STEP] step=79 reward=0.0615 loss=0.0000 kl=0.0142
+[TRAIN_STEP] step=80 reward=0.0490 loss=0.0000 kl=0.0099
+[TRAIN_END] steps_completed=80
+[EVAL_START] phase=trained
+[EVAL] phase=trained workflow=A episode=1 score=0.2290
+[EVAL] phase=trained workflow=A episode=2 score=0.2375
+[EVAL] phase=trained workflow=A episode=3 score=0.2500
+[EVAL] phase=trained workflow=A episode=4 score=0.2500
+[EVAL] phase=trained workflow=A episode=5 score=0.2375
+[EVAL_WORKFLOW] phase=trained workflow=A mean=0.2408 min=0.2290 max=0.2500
+[EVAL] phase=trained workflow=B episode=1 score=0.2500
+[EVAL] phase=trained workflow=B episode=2 score=0.2500
+[EVAL] phase=trained workflow=B episode=3 score=0.2500
+[EVAL] phase=trained workflow=B episode=4 score=0.2500
+[EVAL] phase=trained workflow=B episode=5 score=0.2500
+[EVAL_WORKFLOW] phase=trained workflow=B mean=0.2500 min=0.2500 max=0.2500
+[EVAL] phase=trained workflow=C episode=1 score=0.2230
+[EVAL] phase=trained workflow=C episode=2 score=0.2500
+[EVAL] phase=trained workflow=C episode=3 score=0.0010
+[EVAL] phase=trained workflow=C episode=4 score=0.2470
+[EVAL] phase=trained workflow=C episode=5 score=0.2500
+[EVAL_WORKFLOW] phase=trained workflow=C mean=0.1942 min=0.0010 max=0.2500
+[EVAL_END] phase=trained overall_mean=0.2283
+[TRAIN_SUMMARY] baseline_overall=0.1734 trained_overall=0.2283 delta=+0.0549
+[ARTIFACT] training_curve.png saved
+[ARTIFACT] baseline_vs_trained.png saved
+[ARTIFACT] score_distribution.png saved
+[ARTIFACT] lora_adapter saved to /content/OpenEnv-Round-2/training/orgos_lora_adapter
+[ARTIFACT] lora_adapter saved to /content/OpenEnv-Round-2/training/orgos_lora_adapter