Upload full trainer checkpoint step=4056 (profile=llama3_2_1b_base_grpo, run_type=grpo)
Browse files- .gitattributes +1 -0
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/README.md +210 -0
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/adapter_config.json +50 -0
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/adapter_model.safetensors +3 -0
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/chat_template.jinja +12 -0
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/optimizer.pt +3 -0
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/rng_state.pth +3 -0
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/scheduler.pt +3 -0
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/tokenizer.json +3 -0
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/tokenizer_config.json +18 -0
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/trainer_state.json +2293 -0
- meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/training_args.bin +3 -0
- meta-llama__llama-3.2-1b/grpo/run_manifest.json +8 -2
.gitattributes
CHANGED
|
@@ -106,3 +106,4 @@ meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-2028/tokenizer.json filter=
|
|
| 106 |
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-2535/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 107 |
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-3042/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 108 |
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-3549/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 106 |
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-2535/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 107 |
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-3042/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 108 |
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-3549/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 109 |
+
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/README.md
ADDED
|
@@ -0,0 +1,210 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: unsloth/Llama-3.2-1B
|
| 3 |
+
library_name: peft
|
| 4 |
+
pipeline_tag: text-generation
|
| 5 |
+
tags:
|
| 6 |
+
- base_model:adapter:unsloth/Llama-3.2-1B
|
| 7 |
+
- grpo
|
| 8 |
+
- lora
|
| 9 |
+
- transformers
|
| 10 |
+
- trl
|
| 11 |
+
- unsloth
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# Model Card for Model ID
|
| 15 |
+
|
| 16 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
## Model Details
|
| 21 |
+
|
| 22 |
+
### Model Description
|
| 23 |
+
|
| 24 |
+
<!-- Provide a longer summary of what this model is. -->
|
| 25 |
+
|
| 26 |
+
|
| 27 |
+
|
| 28 |
+
- **Developed by:** [More Information Needed]
|
| 29 |
+
- **Funded by [optional]:** [More Information Needed]
|
| 30 |
+
- **Shared by [optional]:** [More Information Needed]
|
| 31 |
+
- **Model type:** [More Information Needed]
|
| 32 |
+
- **Language(s) (NLP):** [More Information Needed]
|
| 33 |
+
- **License:** [More Information Needed]
|
| 34 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
| 35 |
+
|
| 36 |
+
### Model Sources [optional]
|
| 37 |
+
|
| 38 |
+
<!-- Provide the basic links for the model. -->
|
| 39 |
+
|
| 40 |
+
- **Repository:** [More Information Needed]
|
| 41 |
+
- **Paper [optional]:** [More Information Needed]
|
| 42 |
+
- **Demo [optional]:** [More Information Needed]
|
| 43 |
+
|
| 44 |
+
## Uses
|
| 45 |
+
|
| 46 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 47 |
+
|
| 48 |
+
### Direct Use
|
| 49 |
+
|
| 50 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
| 51 |
+
|
| 52 |
+
[More Information Needed]
|
| 53 |
+
|
| 54 |
+
### Downstream Use [optional]
|
| 55 |
+
|
| 56 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
| 57 |
+
|
| 58 |
+
[More Information Needed]
|
| 59 |
+
|
| 60 |
+
### Out-of-Scope Use
|
| 61 |
+
|
| 62 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 63 |
+
|
| 64 |
+
[More Information Needed]
|
| 65 |
+
|
| 66 |
+
## Bias, Risks, and Limitations
|
| 67 |
+
|
| 68 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 69 |
+
|
| 70 |
+
[More Information Needed]
|
| 71 |
+
|
| 72 |
+
### Recommendations
|
| 73 |
+
|
| 74 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 75 |
+
|
| 76 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
| 77 |
+
|
| 78 |
+
## How to Get Started with the Model
|
| 79 |
+
|
| 80 |
+
Use the code below to get started with the model.
|
| 81 |
+
|
| 82 |
+
[More Information Needed]
|
| 83 |
+
|
| 84 |
+
## Training Details
|
| 85 |
+
|
| 86 |
+
### Training Data
|
| 87 |
+
|
| 88 |
+
<!-- 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. -->
|
| 89 |
+
|
| 90 |
+
[More Information Needed]
|
| 91 |
+
|
| 92 |
+
### Training Procedure
|
| 93 |
+
|
| 94 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
| 95 |
+
|
| 96 |
+
#### Preprocessing [optional]
|
| 97 |
+
|
| 98 |
+
[More Information Needed]
|
| 99 |
+
|
| 100 |
+
|
| 101 |
+
#### Training Hyperparameters
|
| 102 |
+
|
| 103 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
| 104 |
+
|
| 105 |
+
#### Speeds, Sizes, Times [optional]
|
| 106 |
+
|
| 107 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 108 |
+
|
| 109 |
+
[More Information Needed]
|
| 110 |
+
|
| 111 |
+
## Evaluation
|
| 112 |
+
|
| 113 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
| 114 |
+
|
| 115 |
+
### Testing Data, Factors & Metrics
|
| 116 |
+
|
| 117 |
+
#### Testing Data
|
| 118 |
+
|
| 119 |
+
<!-- This should link to a Dataset Card if possible. -->
|
| 120 |
+
|
| 121 |
+
[More Information Needed]
|
| 122 |
+
|
| 123 |
+
#### Factors
|
| 124 |
+
|
| 125 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 126 |
+
|
| 127 |
+
[More Information Needed]
|
| 128 |
+
|
| 129 |
+
#### Metrics
|
| 130 |
+
|
| 131 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
| 132 |
+
|
| 133 |
+
[More Information Needed]
|
| 134 |
+
|
| 135 |
+
### Results
|
| 136 |
+
|
| 137 |
+
[More Information Needed]
|
| 138 |
+
|
| 139 |
+
#### Summary
|
| 140 |
+
|
| 141 |
+
|
| 142 |
+
|
| 143 |
+
## Model Examination [optional]
|
| 144 |
+
|
| 145 |
+
<!-- Relevant interpretability work for the model goes here -->
|
| 146 |
+
|
| 147 |
+
[More Information Needed]
|
| 148 |
+
|
| 149 |
+
## Environmental Impact
|
| 150 |
+
|
| 151 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
| 152 |
+
|
| 153 |
+
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).
|
| 154 |
+
|
| 155 |
+
- **Hardware Type:** [More Information Needed]
|
| 156 |
+
- **Hours used:** [More Information Needed]
|
| 157 |
+
- **Cloud Provider:** [More Information Needed]
|
| 158 |
+
- **Compute Region:** [More Information Needed]
|
| 159 |
+
- **Carbon Emitted:** [More Information Needed]
|
| 160 |
+
|
| 161 |
+
## Technical Specifications [optional]
|
| 162 |
+
|
| 163 |
+
### Model Architecture and Objective
|
| 164 |
+
|
| 165 |
+
[More Information Needed]
|
| 166 |
+
|
| 167 |
+
### Compute Infrastructure
|
| 168 |
+
|
| 169 |
+
[More Information Needed]
|
| 170 |
+
|
| 171 |
+
#### Hardware
|
| 172 |
+
|
| 173 |
+
[More Information Needed]
|
| 174 |
+
|
| 175 |
+
#### Software
|
| 176 |
+
|
| 177 |
+
[More Information Needed]
|
| 178 |
+
|
| 179 |
+
## Citation [optional]
|
| 180 |
+
|
| 181 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 182 |
+
|
| 183 |
+
**BibTeX:**
|
| 184 |
+
|
| 185 |
+
[More Information Needed]
|
| 186 |
+
|
| 187 |
+
**APA:**
|
| 188 |
+
|
| 189 |
+
[More Information Needed]
|
| 190 |
+
|
| 191 |
+
## Glossary [optional]
|
| 192 |
+
|
| 193 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 194 |
+
|
| 195 |
+
[More Information Needed]
|
| 196 |
+
|
| 197 |
+
## More Information [optional]
|
| 198 |
+
|
| 199 |
+
[More Information Needed]
|
| 200 |
+
|
| 201 |
+
## Model Card Authors [optional]
|
| 202 |
+
|
| 203 |
+
[More Information Needed]
|
| 204 |
+
|
| 205 |
+
## Model Card Contact
|
| 206 |
+
|
| 207 |
+
[More Information Needed]
|
| 208 |
+
### Framework versions
|
| 209 |
+
|
| 210 |
+
- PEFT 0.18.1
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/adapter_config.json
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {},
|
| 4 |
+
"arrow_config": null,
|
| 5 |
+
"auto_mapping": {
|
| 6 |
+
"base_model_class": "LlamaForCausalLM",
|
| 7 |
+
"parent_library": "transformers.models.llama.modeling_llama",
|
| 8 |
+
"unsloth_fixed": true
|
| 9 |
+
},
|
| 10 |
+
"base_model_name_or_path": "unsloth/Llama-3.2-1B",
|
| 11 |
+
"bias": "none",
|
| 12 |
+
"corda_config": null,
|
| 13 |
+
"ensure_weight_tying": false,
|
| 14 |
+
"eva_config": null,
|
| 15 |
+
"exclude_modules": null,
|
| 16 |
+
"fan_in_fan_out": false,
|
| 17 |
+
"inference_mode": true,
|
| 18 |
+
"init_lora_weights": true,
|
| 19 |
+
"layer_replication": null,
|
| 20 |
+
"layers_pattern": null,
|
| 21 |
+
"layers_to_transform": null,
|
| 22 |
+
"loftq_config": {},
|
| 23 |
+
"lora_alpha": 16,
|
| 24 |
+
"lora_bias": false,
|
| 25 |
+
"lora_dropout": 0.0,
|
| 26 |
+
"megatron_config": null,
|
| 27 |
+
"megatron_core": "megatron.core",
|
| 28 |
+
"modules_to_save": null,
|
| 29 |
+
"peft_type": "LORA",
|
| 30 |
+
"peft_version": "0.18.1",
|
| 31 |
+
"qalora_group_size": 16,
|
| 32 |
+
"r": 16,
|
| 33 |
+
"rank_pattern": {},
|
| 34 |
+
"revision": null,
|
| 35 |
+
"target_modules": [
|
| 36 |
+
"down_proj",
|
| 37 |
+
"o_proj",
|
| 38 |
+
"q_proj",
|
| 39 |
+
"gate_proj",
|
| 40 |
+
"k_proj",
|
| 41 |
+
"up_proj",
|
| 42 |
+
"v_proj"
|
| 43 |
+
],
|
| 44 |
+
"target_parameters": null,
|
| 45 |
+
"task_type": "CAUSAL_LM",
|
| 46 |
+
"trainable_token_indices": null,
|
| 47 |
+
"use_dora": false,
|
| 48 |
+
"use_qalora": false,
|
| 49 |
+
"use_rslora": false
|
| 50 |
+
}
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:69d81296ae1ef03a4797ad3128e9764ea766bbf99329940a78280208a4eea4c8
|
| 3 |
+
size 45118424
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/chat_template.jinja
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ messages[0]['content'] + '
|
| 2 |
+
|
| 3 |
+
' }}{% set loop_messages = messages[1:] %}{% else %}{{ '' + '
|
| 4 |
+
|
| 5 |
+
' }}{% set loop_messages = messages %}{% endif %}{% for message in loop_messages %}{% if message['role'] == 'user' %}{{ '### Instruction:
|
| 6 |
+
' + message['content'] + '
|
| 7 |
+
|
| 8 |
+
' }}{% elif message['role'] == 'assistant' %}{{ '### Response:
|
| 9 |
+
' + message['content'] + eos_token + '
|
| 10 |
+
|
| 11 |
+
' }}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '### Response:
|
| 12 |
+
' }}{% endif %}
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:47970fdc51c4cd479bc84b0d37d7d85c4bd37cba379d70d58a2934fbb88dbb5c
|
| 3 |
+
size 23162187
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3f8c469f4824664eebfa711a55bf59c3dde9210e846024edc02ca5f310c9e388
|
| 3 |
+
size 14645
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6286f1b48597a0884f0d88baaf6c3aa0ea3eb26423fd7758be35f5e8ad235bd6
|
| 3 |
+
size 1465
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a65c6c5f9764771aa485e6a1f5e63d7d9af8477fe0777148c17476ecb2e09a05
|
| 3 |
+
size 17210099
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/tokenizer_config.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|begin_of_text|>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|end_of_text|>",
|
| 6 |
+
"from_slow": true,
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"legacy": false,
|
| 9 |
+
"model_input_names": [
|
| 10 |
+
"input_ids",
|
| 11 |
+
"attention_mask"
|
| 12 |
+
],
|
| 13 |
+
"model_max_length": 131072,
|
| 14 |
+
"pad_token": "<|finetune_right_pad_id|>",
|
| 15 |
+
"padding_side": "right",
|
| 16 |
+
"tokenizer_class": "TokenizersBackend",
|
| 17 |
+
"unk_token": null
|
| 18 |
+
}
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/trainer_state.json
ADDED
|
@@ -0,0 +1,2293 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.007204160509264522,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 4056,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"clip_ratio/high_max": 0.0,
|
| 14 |
+
"clip_ratio/high_mean": 0.0,
|
| 15 |
+
"clip_ratio/low_mean": 0.0,
|
| 16 |
+
"clip_ratio/low_min": 0.0,
|
| 17 |
+
"clip_ratio/region_mean": 0.0,
|
| 18 |
+
"completion_length": 65.67,
|
| 19 |
+
"completions/clipped_ratio": 0.0,
|
| 20 |
+
"completions/max_length": 93.75,
|
| 21 |
+
"completions/max_terminated_length": 93.75,
|
| 22 |
+
"completions/mean_length": 63.3203125,
|
| 23 |
+
"completions/mean_terminated_length": 63.3203125,
|
| 24 |
+
"completions/min_length": 38.25,
|
| 25 |
+
"completions/min_terminated_length": 38.25,
|
| 26 |
+
"epoch": 8.88086847789019e-05,
|
| 27 |
+
"frac_reward_zero_std": 0.5625,
|
| 28 |
+
"grad_norm": 0.007002627942711115,
|
| 29 |
+
"kl": 0.0,
|
| 30 |
+
"learning_rate": 2.7197442330321267e-08,
|
| 31 |
+
"loss": -0.0006554330885410309,
|
| 32 |
+
"num_tokens": 57618.0,
|
| 33 |
+
"reward": -6.51925802230835e-09,
|
| 34 |
+
"reward_std": 0.48731958121061325,
|
| 35 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -8.381903171539307e-09,
|
| 36 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.007905274629593,
|
| 37 |
+
"step": 50
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"clip_ratio/high_max": 0.0,
|
| 41 |
+
"clip_ratio/high_mean": 0.0,
|
| 42 |
+
"clip_ratio/low_mean": 0.0,
|
| 43 |
+
"clip_ratio/low_min": 0.0,
|
| 44 |
+
"clip_ratio/region_mean": 0.0,
|
| 45 |
+
"completion_length": 52.34,
|
| 46 |
+
"completions/clipped_ratio": 0.0,
|
| 47 |
+
"completions/max_length": 84.33333333333333,
|
| 48 |
+
"completions/max_terminated_length": 84.33333333333333,
|
| 49 |
+
"completions/mean_length": 51.208333333333336,
|
| 50 |
+
"completions/mean_terminated_length": 51.208333333333336,
|
| 51 |
+
"completions/min_length": 26.0,
|
| 52 |
+
"completions/min_terminated_length": 26.0,
|
| 53 |
+
"epoch": 0.0001776173695578038,
|
| 54 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 55 |
+
"grad_norm": 0.0,
|
| 56 |
+
"kl": 0.0,
|
| 57 |
+
"learning_rate": 5.494993450411847e-08,
|
| 58 |
+
"loss": 0.0013675823807716369,
|
| 59 |
+
"num_tokens": 94074.0,
|
| 60 |
+
"reward": 1.4901161193847656e-08,
|
| 61 |
+
"reward_std": 0.44680649042129517,
|
| 62 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.179675261179606e-08,
|
| 63 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052845637004,
|
| 64 |
+
"step": 100
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"clip_ratio/high_max": 0.0,
|
| 68 |
+
"clip_ratio/high_mean": 0.0,
|
| 69 |
+
"clip_ratio/low_mean": 0.00017899675294756888,
|
| 70 |
+
"clip_ratio/low_min": 0.00017899675294756888,
|
| 71 |
+
"clip_ratio/region_mean": 0.00017899675294756888,
|
| 72 |
+
"completion_length": 64.945,
|
| 73 |
+
"completions/clipped_ratio": 0.0,
|
| 74 |
+
"completions/max_length": 119.33333333333333,
|
| 75 |
+
"completions/max_terminated_length": 119.33333333333333,
|
| 76 |
+
"completions/mean_length": 66.65625,
|
| 77 |
+
"completions/mean_terminated_length": 66.65625,
|
| 78 |
+
"completions/min_length": 39.0,
|
| 79 |
+
"completions/min_terminated_length": 39.0,
|
| 80 |
+
"epoch": 0.0002664260543367057,
|
| 81 |
+
"frac_reward_zero_std": 0.7083333333333334,
|
| 82 |
+
"grad_norm": 0.015731461346149445,
|
| 83 |
+
"kl": 0.0,
|
| 84 |
+
"learning_rate": 8.270242667791569e-08,
|
| 85 |
+
"loss": 0.0017325745522975922,
|
| 86 |
+
"num_tokens": 139152.0,
|
| 87 |
+
"reward": 2.483526865641276e-09,
|
| 88 |
+
"reward_std": 0.2613494098186493,
|
| 89 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 0.0,
|
| 90 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 91 |
+
"step": 150
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"clip_ratio/high_max": 0.0,
|
| 95 |
+
"clip_ratio/high_mean": 0.0,
|
| 96 |
+
"clip_ratio/low_mean": 0.0,
|
| 97 |
+
"clip_ratio/low_min": 0.0,
|
| 98 |
+
"clip_ratio/region_mean": 0.0,
|
| 99 |
+
"completion_length": 60.15,
|
| 100 |
+
"completions/clipped_ratio": 0.0,
|
| 101 |
+
"completions/max_length": 99.0,
|
| 102 |
+
"completions/max_terminated_length": 99.0,
|
| 103 |
+
"completions/mean_length": 63.015625,
|
| 104 |
+
"completions/mean_terminated_length": 63.015625,
|
| 105 |
+
"completions/min_length": 32.666666666666664,
|
| 106 |
+
"completions/min_terminated_length": 32.666666666666664,
|
| 107 |
+
"epoch": 0.0003552347391156076,
|
| 108 |
+
"frac_reward_zero_std": 0.4583333333333333,
|
| 109 |
+
"grad_norm": 0.039579447358846664,
|
| 110 |
+
"kl": 0.0,
|
| 111 |
+
"learning_rate": 1.1045491885171288e-07,
|
| 112 |
+
"loss": -0.0044269835948944096,
|
| 113 |
+
"num_tokens": 182803.0,
|
| 114 |
+
"reward": -4.967053731282552e-09,
|
| 115 |
+
"reward_std": 0.5628267228603363,
|
| 116 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -2.483526865641276e-09,
|
| 117 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052845637004,
|
| 118 |
+
"step": 200
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"clip_ratio/high_max": 0.0,
|
| 122 |
+
"clip_ratio/high_mean": 0.0,
|
| 123 |
+
"clip_ratio/low_mean": 0.00016559829469770193,
|
| 124 |
+
"clip_ratio/low_min": 0.00016559829469770193,
|
| 125 |
+
"clip_ratio/region_mean": 0.00016559829469770193,
|
| 126 |
+
"completion_length": 64.905,
|
| 127 |
+
"completions/clipped_ratio": 0.0,
|
| 128 |
+
"completions/max_length": 98.66666666666667,
|
| 129 |
+
"completions/max_terminated_length": 98.66666666666667,
|
| 130 |
+
"completions/mean_length": 64.203125,
|
| 131 |
+
"completions/mean_terminated_length": 64.203125,
|
| 132 |
+
"completions/min_length": 35.666666666666664,
|
| 133 |
+
"completions/min_terminated_length": 35.666666666666664,
|
| 134 |
+
"epoch": 0.0004440434238945095,
|
| 135 |
+
"frac_reward_zero_std": 0.4166666666666667,
|
| 136 |
+
"grad_norm": 0.011699045076966286,
|
| 137 |
+
"kl": 0.0,
|
| 138 |
+
"learning_rate": 1.382074110255101e-07,
|
| 139 |
+
"loss": 0.00223430335521698,
|
| 140 |
+
"num_tokens": 227314.0,
|
| 141 |
+
"reward": 2.2351741790771484e-08,
|
| 142 |
+
"reward_std": 0.4298504690329234,
|
| 143 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.241763432820638e-08,
|
| 144 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 145 |
+
"step": 250
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"clip_ratio/high_max": 0.0,
|
| 149 |
+
"clip_ratio/high_mean": 0.0,
|
| 150 |
+
"clip_ratio/low_mean": 0.0,
|
| 151 |
+
"clip_ratio/low_min": 0.0,
|
| 152 |
+
"clip_ratio/region_mean": 0.0,
|
| 153 |
+
"completion_length": 70.07,
|
| 154 |
+
"completions/clipped_ratio": 0.0,
|
| 155 |
+
"completions/max_length": 108.0,
|
| 156 |
+
"completions/max_terminated_length": 108.0,
|
| 157 |
+
"completions/mean_length": 70.63541666666667,
|
| 158 |
+
"completions/mean_terminated_length": 70.63541666666667,
|
| 159 |
+
"completions/min_length": 34.0,
|
| 160 |
+
"completions/min_terminated_length": 34.0,
|
| 161 |
+
"epoch": 0.0005328521086734114,
|
| 162 |
+
"frac_reward_zero_std": 0.5833333333333334,
|
| 163 |
+
"grad_norm": 0.0456908717751503,
|
| 164 |
+
"kl": 0.0,
|
| 165 |
+
"learning_rate": 1.659599031993073e-07,
|
| 166 |
+
"loss": -0.0011175717413425446,
|
| 167 |
+
"num_tokens": 275956.0,
|
| 168 |
+
"reward": 5.587935447692871e-09,
|
| 169 |
+
"reward_std": 0.4012269576390584,
|
| 170 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 3.104408582051595e-09,
|
| 171 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 172 |
+
"step": 300
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"clip_ratio/high_max": 0.0,
|
| 176 |
+
"clip_ratio/high_mean": 0.0,
|
| 177 |
+
"clip_ratio/low_mean": 5.8139534667134284e-05,
|
| 178 |
+
"clip_ratio/low_min": 5.8139534667134284e-05,
|
| 179 |
+
"clip_ratio/region_mean": 5.8139534667134284e-05,
|
| 180 |
+
"completion_length": 62.93,
|
| 181 |
+
"completions/clipped_ratio": 0.0,
|
| 182 |
+
"completions/max_length": 95.66666666666667,
|
| 183 |
+
"completions/max_terminated_length": 95.66666666666667,
|
| 184 |
+
"completions/mean_length": 61.6875,
|
| 185 |
+
"completions/mean_terminated_length": 61.6875,
|
| 186 |
+
"completions/min_length": 29.333333333333332,
|
| 187 |
+
"completions/min_terminated_length": 29.333333333333332,
|
| 188 |
+
"epoch": 0.0006216607934523133,
|
| 189 |
+
"frac_reward_zero_std": 0.75,
|
| 190 |
+
"grad_norm": 0.0190290417522192,
|
| 191 |
+
"kl": 0.0,
|
| 192 |
+
"learning_rate": 1.9371239537310452e-07,
|
| 193 |
+
"loss": -0.0010719958692789078,
|
| 194 |
+
"num_tokens": 319328.0,
|
| 195 |
+
"reward": -2.483526865641276e-09,
|
| 196 |
+
"reward_std": 0.34829530119895935,
|
| 197 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -4.967053731282552e-09,
|
| 198 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 199 |
+
"step": 350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"clip_ratio/high_max": 0.0,
|
| 203 |
+
"clip_ratio/high_mean": 0.0,
|
| 204 |
+
"clip_ratio/low_mean": 7.352941203862428e-05,
|
| 205 |
+
"clip_ratio/low_min": 7.352941203862428e-05,
|
| 206 |
+
"clip_ratio/region_mean": 7.352941203862428e-05,
|
| 207 |
+
"completion_length": 62.725,
|
| 208 |
+
"completions/clipped_ratio": 0.0,
|
| 209 |
+
"completions/max_length": 105.66666666666667,
|
| 210 |
+
"completions/max_terminated_length": 105.66666666666667,
|
| 211 |
+
"completions/mean_length": 62.890625,
|
| 212 |
+
"completions/mean_terminated_length": 62.890625,
|
| 213 |
+
"completions/min_length": 34.666666666666664,
|
| 214 |
+
"completions/min_terminated_length": 34.666666666666664,
|
| 215 |
+
"epoch": 0.0007104694782312152,
|
| 216 |
+
"frac_reward_zero_std": 0.5,
|
| 217 |
+
"grad_norm": 0.02625960297882557,
|
| 218 |
+
"kl": 0.0,
|
| 219 |
+
"learning_rate": 2.2146488754690174e-07,
|
| 220 |
+
"loss": 0.0013731226325035095,
|
| 221 |
+
"num_tokens": 362563.0,
|
| 222 |
+
"reward": 4.967053731282552e-09,
|
| 223 |
+
"reward_std": 0.44316299756368,
|
| 224 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 0.0,
|
| 225 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 226 |
+
"step": 400
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"clip_ratio/high_max": 0.0,
|
| 230 |
+
"clip_ratio/high_mean": 0.0,
|
| 231 |
+
"clip_ratio/low_mean": 8.695651777088643e-05,
|
| 232 |
+
"clip_ratio/low_min": 8.695651777088643e-05,
|
| 233 |
+
"clip_ratio/region_mean": 8.695651777088643e-05,
|
| 234 |
+
"completion_length": 66.03,
|
| 235 |
+
"completions/clipped_ratio": 0.0,
|
| 236 |
+
"completions/max_length": 112.0,
|
| 237 |
+
"completions/max_terminated_length": 112.0,
|
| 238 |
+
"completions/mean_length": 70.2890625,
|
| 239 |
+
"completions/mean_terminated_length": 70.2890625,
|
| 240 |
+
"completions/min_length": 42.75,
|
| 241 |
+
"completions/min_terminated_length": 42.75,
|
| 242 |
+
"epoch": 0.0007992781630101171,
|
| 243 |
+
"frac_reward_zero_std": 0.5625,
|
| 244 |
+
"grad_norm": 0.0035960455425083637,
|
| 245 |
+
"kl": 0.0,
|
| 246 |
+
"learning_rate": 2.492173797206989e-07,
|
| 247 |
+
"loss": -0.0004108186066150665,
|
| 248 |
+
"num_tokens": 427021.0,
|
| 249 |
+
"reward": 1.30385160446167e-08,
|
| 250 |
+
"reward_std": 0.383321788161993,
|
| 251 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.4901161193847656e-08,
|
| 252 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 253 |
+
"step": 450
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"clip_ratio/high_max": 0.0,
|
| 257 |
+
"clip_ratio/high_mean": 0.0,
|
| 258 |
+
"clip_ratio/low_mean": 0.0,
|
| 259 |
+
"clip_ratio/low_min": 0.0,
|
| 260 |
+
"clip_ratio/region_mean": 0.0,
|
| 261 |
+
"completion_length": 63.98,
|
| 262 |
+
"completions/clipped_ratio": 0.0,
|
| 263 |
+
"completions/max_length": 85.33333333333333,
|
| 264 |
+
"completions/max_terminated_length": 85.33333333333333,
|
| 265 |
+
"completions/mean_length": 55.046875,
|
| 266 |
+
"completions/mean_terminated_length": 55.046875,
|
| 267 |
+
"completions/min_length": 29.666666666666668,
|
| 268 |
+
"completions/min_terminated_length": 29.666666666666668,
|
| 269 |
+
"epoch": 0.000888086847789019,
|
| 270 |
+
"frac_reward_zero_std": 0.7916666666666666,
|
| 271 |
+
"grad_norm": 0.0006704159895889461,
|
| 272 |
+
"kl": 0.0,
|
| 273 |
+
"learning_rate": 2.7696987189449615e-07,
|
| 274 |
+
"loss": 0.0006217561289668084,
|
| 275 |
+
"num_tokens": 464766.0,
|
| 276 |
+
"reward": -4.967053731282552e-09,
|
| 277 |
+
"reward_std": 0.1608196645975113,
|
| 278 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 4.967053731282552e-09,
|
| 279 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 0.6719369093577067,
|
| 280 |
+
"step": 500
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"eval_cer_subset": 0.013239532952269286,
|
| 284 |
+
"eval_cer_subset_edit_distance": 813,
|
| 285 |
+
"eval_cer_subset_groups": 250,
|
| 286 |
+
"eval_cer_subset_items": 250,
|
| 287 |
+
"eval_cer_subset_ref_chars": 61407,
|
| 288 |
+
"eval_loss": 0.03968957871396896,
|
| 289 |
+
"step": 507
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"clip_ratio/high_max": 0.00010526316240429879,
|
| 293 |
+
"clip_ratio/high_mean": 0.00010526316240429879,
|
| 294 |
+
"clip_ratio/low_mean": 5.917159840464592e-05,
|
| 295 |
+
"clip_ratio/low_min": 5.917159840464592e-05,
|
| 296 |
+
"clip_ratio/region_mean": 0.0001644347608089447,
|
| 297 |
+
"completion_length": 65.61,
|
| 298 |
+
"completions/clipped_ratio": 0.0,
|
| 299 |
+
"completions/max_length": 114.66666666666667,
|
| 300 |
+
"completions/max_terminated_length": 114.66666666666667,
|
| 301 |
+
"completions/mean_length": 71.984375,
|
| 302 |
+
"completions/mean_terminated_length": 71.984375,
|
| 303 |
+
"completions/min_length": 37.0,
|
| 304 |
+
"completions/min_terminated_length": 37.0,
|
| 305 |
+
"epoch": 0.000976895532567921,
|
| 306 |
+
"frac_reward_zero_std": 0.7083333333333334,
|
| 307 |
+
"grad_norm": 0.008066101931035519,
|
| 308 |
+
"kl": 0.0,
|
| 309 |
+
"learning_rate": 3.0472236406829335e-07,
|
| 310 |
+
"loss": -3.333181142807007e-05,
|
| 311 |
+
"num_tokens": 513963.0,
|
| 312 |
+
"reward": 0.0,
|
| 313 |
+
"reward_std": 0.36842172344525653,
|
| 314 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 0.0,
|
| 315 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 316 |
+
"step": 550
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"clip_ratio/high_max": 0.0,
|
| 320 |
+
"clip_ratio/high_mean": 0.0,
|
| 321 |
+
"clip_ratio/low_mean": 9.615384973585606e-05,
|
| 322 |
+
"clip_ratio/low_min": 9.615384973585606e-05,
|
| 323 |
+
"clip_ratio/region_mean": 9.615384973585606e-05,
|
| 324 |
+
"completion_length": 73.1,
|
| 325 |
+
"completions/clipped_ratio": 0.0,
|
| 326 |
+
"completions/max_length": 108.33333333333333,
|
| 327 |
+
"completions/max_terminated_length": 108.33333333333333,
|
| 328 |
+
"completions/mean_length": 71.40104166666667,
|
| 329 |
+
"completions/mean_terminated_length": 71.40104166666667,
|
| 330 |
+
"completions/min_length": 47.0,
|
| 331 |
+
"completions/min_terminated_length": 47.0,
|
| 332 |
+
"epoch": 0.0010657042173468228,
|
| 333 |
+
"frac_reward_zero_std": 0.4583333333333333,
|
| 334 |
+
"grad_norm": 0.00526217557489872,
|
| 335 |
+
"kl": 0.0,
|
| 336 |
+
"learning_rate": 3.3247485624209054e-07,
|
| 337 |
+
"loss": 0.00033511649817228315,
|
| 338 |
+
"num_tokens": 562392.0,
|
| 339 |
+
"reward": -4.967053731282552e-09,
|
| 340 |
+
"reward_std": 0.35033048689365387,
|
| 341 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -2.483526865641276e-09,
|
| 342 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 343 |
+
"step": 600
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"clip_ratio/high_max": 0.0,
|
| 347 |
+
"clip_ratio/high_mean": 0.0,
|
| 348 |
+
"clip_ratio/low_mean": 6.493506487458944e-05,
|
| 349 |
+
"clip_ratio/low_min": 6.493506487458944e-05,
|
| 350 |
+
"clip_ratio/region_mean": 6.493506487458944e-05,
|
| 351 |
+
"completion_length": 71.48,
|
| 352 |
+
"completions/clipped_ratio": 0.0,
|
| 353 |
+
"completions/max_length": 111.66666666666667,
|
| 354 |
+
"completions/max_terminated_length": 111.66666666666667,
|
| 355 |
+
"completions/mean_length": 71.27604166666667,
|
| 356 |
+
"completions/mean_terminated_length": 71.27604166666667,
|
| 357 |
+
"completions/min_length": 39.666666666666664,
|
| 358 |
+
"completions/min_terminated_length": 39.666666666666664,
|
| 359 |
+
"epoch": 0.0011545129021257247,
|
| 360 |
+
"frac_reward_zero_std": 0.5416666666666666,
|
| 361 |
+
"grad_norm": 0.02180999144911766,
|
| 362 |
+
"kl": 0.0,
|
| 363 |
+
"learning_rate": 3.602273484158878e-07,
|
| 364 |
+
"loss": -0.0010874558240175248,
|
| 365 |
+
"num_tokens": 610477.0,
|
| 366 |
+
"reward": -4.967053731282552e-09,
|
| 367 |
+
"reward_std": 0.29229696094989777,
|
| 368 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -4.967053731282552e-09,
|
| 369 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 370 |
+
"step": 650
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"clip_ratio/high_max": 0.0,
|
| 374 |
+
"clip_ratio/high_mean": 0.0,
|
| 375 |
+
"clip_ratio/low_mean": 0.0,
|
| 376 |
+
"clip_ratio/low_min": 0.0,
|
| 377 |
+
"clip_ratio/region_mean": 0.0,
|
| 378 |
+
"completion_length": 68.335,
|
| 379 |
+
"completions/clipped_ratio": 0.0,
|
| 380 |
+
"completions/max_length": 106.66666666666667,
|
| 381 |
+
"completions/max_terminated_length": 106.66666666666667,
|
| 382 |
+
"completions/mean_length": 67.78645833333333,
|
| 383 |
+
"completions/mean_terminated_length": 67.78645833333333,
|
| 384 |
+
"completions/min_length": 34.0,
|
| 385 |
+
"completions/min_terminated_length": 34.0,
|
| 386 |
+
"epoch": 0.0012433215869046266,
|
| 387 |
+
"frac_reward_zero_std": 0.5,
|
| 388 |
+
"grad_norm": 0.03915110602974892,
|
| 389 |
+
"kl": 0.0,
|
| 390 |
+
"learning_rate": 3.87979840589685e-07,
|
| 391 |
+
"loss": -0.0011869536340236664,
|
| 392 |
+
"num_tokens": 657028.0,
|
| 393 |
+
"reward": 2.483526865641276e-09,
|
| 394 |
+
"reward_std": 0.5572122434775034,
|
| 395 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.483526865641276e-09,
|
| 396 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 397 |
+
"step": 700
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"clip_ratio/high_max": 0.0,
|
| 401 |
+
"clip_ratio/high_mean": 0.0,
|
| 402 |
+
"clip_ratio/low_mean": 0.0,
|
| 403 |
+
"clip_ratio/low_min": 0.0,
|
| 404 |
+
"clip_ratio/region_mean": 0.0,
|
| 405 |
+
"completion_length": 67.125,
|
| 406 |
+
"completions/clipped_ratio": 0.0,
|
| 407 |
+
"completions/max_length": 105.33333333333333,
|
| 408 |
+
"completions/max_terminated_length": 105.33333333333333,
|
| 409 |
+
"completions/mean_length": 67.28645833333333,
|
| 410 |
+
"completions/mean_terminated_length": 67.28645833333333,
|
| 411 |
+
"completions/min_length": 35.333333333333336,
|
| 412 |
+
"completions/min_terminated_length": 35.333333333333336,
|
| 413 |
+
"epoch": 0.0013321302716835285,
|
| 414 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 415 |
+
"grad_norm": 0.048128049820661545,
|
| 416 |
+
"kl": 0.0,
|
| 417 |
+
"learning_rate": 4.1573233276348217e-07,
|
| 418 |
+
"loss": 0.0027270379662513735,
|
| 419 |
+
"num_tokens": 702891.0,
|
| 420 |
+
"reward": 2.483526865641276e-09,
|
| 421 |
+
"reward_std": 0.4348108967145284,
|
| 422 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -6.829698880513509e-09,
|
| 423 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 424 |
+
"step": 750
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"clip_ratio/high_max": 0.0,
|
| 428 |
+
"clip_ratio/high_mean": 0.0,
|
| 429 |
+
"clip_ratio/low_mean": 0.00010526316240429879,
|
| 430 |
+
"clip_ratio/low_min": 0.00010526316240429879,
|
| 431 |
+
"clip_ratio/region_mean": 0.00010526316240429879,
|
| 432 |
+
"completion_length": 67.81,
|
| 433 |
+
"completions/clipped_ratio": 0.0,
|
| 434 |
+
"completions/max_length": 98.66666666666667,
|
| 435 |
+
"completions/max_terminated_length": 98.66666666666667,
|
| 436 |
+
"completions/mean_length": 67.61458333333333,
|
| 437 |
+
"completions/mean_terminated_length": 67.61458333333333,
|
| 438 |
+
"completions/min_length": 34.333333333333336,
|
| 439 |
+
"completions/min_terminated_length": 34.333333333333336,
|
| 440 |
+
"epoch": 0.0014209389564624304,
|
| 441 |
+
"frac_reward_zero_std": 0.5833333333333334,
|
| 442 |
+
"grad_norm": 0.028703227639198303,
|
| 443 |
+
"kl": 0.0,
|
| 444 |
+
"learning_rate": 4.434848249372794e-07,
|
| 445 |
+
"loss": -0.0013799819350242615,
|
| 446 |
+
"num_tokens": 749081.0,
|
| 447 |
+
"reward": -1.4901161193847656e-08,
|
| 448 |
+
"reward_std": 0.40635913610458374,
|
| 449 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -1.4901161193847656e-08,
|
| 450 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 451 |
+
"step": 800
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"clip_ratio/high_max": 0.0,
|
| 455 |
+
"clip_ratio/high_mean": 0.0,
|
| 456 |
+
"clip_ratio/low_mean": 8.333333767950535e-05,
|
| 457 |
+
"clip_ratio/low_min": 8.333333767950535e-05,
|
| 458 |
+
"clip_ratio/region_mean": 8.333333767950535e-05,
|
| 459 |
+
"completion_length": 73.535,
|
| 460 |
+
"completions/clipped_ratio": 0.0,
|
| 461 |
+
"completions/max_length": 113.5,
|
| 462 |
+
"completions/max_terminated_length": 113.5,
|
| 463 |
+
"completions/mean_length": 75.88671875,
|
| 464 |
+
"completions/mean_terminated_length": 75.88671875,
|
| 465 |
+
"completions/min_length": 36.5,
|
| 466 |
+
"completions/min_terminated_length": 36.5,
|
| 467 |
+
"epoch": 0.0015097476412413323,
|
| 468 |
+
"frac_reward_zero_std": 0.59375,
|
| 469 |
+
"grad_norm": 0.002819702262058854,
|
| 470 |
+
"kl": 0.0,
|
| 471 |
+
"learning_rate": 4.712373171110766e-07,
|
| 472 |
+
"loss": -0.0007266353815793992,
|
| 473 |
+
"num_tokens": 818284.0,
|
| 474 |
+
"reward": 9.313225746154785e-09,
|
| 475 |
+
"reward_std": 0.472036711871624,
|
| 476 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.1175870895385742e-08,
|
| 477 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.007905274629593,
|
| 478 |
+
"step": 850
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"clip_ratio/high_max": 4.291845485568046e-05,
|
| 482 |
+
"clip_ratio/high_mean": 4.291845485568046e-05,
|
| 483 |
+
"clip_ratio/low_mean": 4.32900432497263e-05,
|
| 484 |
+
"clip_ratio/low_min": 4.32900432497263e-05,
|
| 485 |
+
"clip_ratio/region_mean": 8.620849810540676e-05,
|
| 486 |
+
"completion_length": 76.735,
|
| 487 |
+
"completions/clipped_ratio": 0.0,
|
| 488 |
+
"completions/max_length": 123.0,
|
| 489 |
+
"completions/max_terminated_length": 123.0,
|
| 490 |
+
"completions/mean_length": 73.05208333333333,
|
| 491 |
+
"completions/mean_terminated_length": 73.05208333333333,
|
| 492 |
+
"completions/min_length": 38.666666666666664,
|
| 493 |
+
"completions/min_terminated_length": 38.666666666666664,
|
| 494 |
+
"epoch": 0.0015985563260202341,
|
| 495 |
+
"frac_reward_zero_std": 0.5416666666666666,
|
| 496 |
+
"grad_norm": 0.0,
|
| 497 |
+
"kl": 0.0,
|
| 498 |
+
"learning_rate": 4.989898092848738e-07,
|
| 499 |
+
"loss": 0.0014878523349761963,
|
| 500 |
+
"num_tokens": 868206.0,
|
| 501 |
+
"reward": 1.241763432820638e-09,
|
| 502 |
+
"reward_std": 0.4125714997450511,
|
| 503 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 4.967053731282552e-09,
|
| 504 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052050908406,
|
| 505 |
+
"step": 900
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"clip_ratio/high_max": 0.0,
|
| 509 |
+
"clip_ratio/high_mean": 0.0,
|
| 510 |
+
"clip_ratio/low_mean": 5.917159840464592e-05,
|
| 511 |
+
"clip_ratio/low_min": 5.917159840464592e-05,
|
| 512 |
+
"clip_ratio/region_mean": 5.917159840464592e-05,
|
| 513 |
+
"completion_length": 69.03,
|
| 514 |
+
"completions/clipped_ratio": 0.0,
|
| 515 |
+
"completions/max_length": 106.33333333333333,
|
| 516 |
+
"completions/max_terminated_length": 106.33333333333333,
|
| 517 |
+
"completions/mean_length": 66.10416666666667,
|
| 518 |
+
"completions/mean_terminated_length": 66.10416666666667,
|
| 519 |
+
"completions/min_length": 36.666666666666664,
|
| 520 |
+
"completions/min_terminated_length": 36.666666666666664,
|
| 521 |
+
"epoch": 0.001687365010799136,
|
| 522 |
+
"frac_reward_zero_std": 0.625,
|
| 523 |
+
"grad_norm": 0.0075177694670856,
|
| 524 |
+
"kl": 0.0,
|
| 525 |
+
"learning_rate": 5.26742301458671e-07,
|
| 526 |
+
"loss": -0.002387867271900177,
|
| 527 |
+
"num_tokens": 913738.0,
|
| 528 |
+
"reward": 0.0,
|
| 529 |
+
"reward_std": 0.4860446055730184,
|
| 530 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.483526865641276e-09,
|
| 531 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 532 |
+
"step": 950
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"clip_ratio/high_max": 0.0,
|
| 536 |
+
"clip_ratio/high_mean": 0.0,
|
| 537 |
+
"clip_ratio/low_mean": 0.0,
|
| 538 |
+
"clip_ratio/low_min": 0.0,
|
| 539 |
+
"clip_ratio/region_mean": 0.0,
|
| 540 |
+
"completion_length": 63.8,
|
| 541 |
+
"completions/clipped_ratio": 0.0,
|
| 542 |
+
"completions/max_length": 106.33333333333333,
|
| 543 |
+
"completions/max_terminated_length": 106.33333333333333,
|
| 544 |
+
"completions/mean_length": 65.10416666666667,
|
| 545 |
+
"completions/mean_terminated_length": 65.10416666666667,
|
| 546 |
+
"completions/min_length": 28.0,
|
| 547 |
+
"completions/min_terminated_length": 28.0,
|
| 548 |
+
"epoch": 0.001776173695578038,
|
| 549 |
+
"frac_reward_zero_std": 0.5416666666666666,
|
| 550 |
+
"grad_norm": 0.02761891484260559,
|
| 551 |
+
"kl": 0.0,
|
| 552 |
+
"learning_rate": 5.544947936324682e-07,
|
| 553 |
+
"loss": 0.0017396166920661926,
|
| 554 |
+
"num_tokens": 958582.0,
|
| 555 |
+
"reward": 0.0,
|
| 556 |
+
"reward_std": 0.4594339330991109,
|
| 557 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.483526865641276e-09,
|
| 558 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079053243001301,
|
| 559 |
+
"step": 1000
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"eval_cer_subset": 0.013044115491719185,
|
| 563 |
+
"eval_cer_subset_edit_distance": 801,
|
| 564 |
+
"eval_cer_subset_groups": 250,
|
| 565 |
+
"eval_cer_subset_items": 250,
|
| 566 |
+
"eval_cer_subset_ref_chars": 61407,
|
| 567 |
+
"eval_loss": 0.03965790307253722,
|
| 568 |
+
"step": 1014
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"clip_ratio/high_max": 0.0,
|
| 572 |
+
"clip_ratio/high_mean": 0.0,
|
| 573 |
+
"clip_ratio/low_mean": 7.633587811142206e-05,
|
| 574 |
+
"clip_ratio/low_min": 7.633587811142206e-05,
|
| 575 |
+
"clip_ratio/region_mean": 7.633587811142206e-05,
|
| 576 |
+
"completion_length": 60.39,
|
| 577 |
+
"completions/clipped_ratio": 0.0,
|
| 578 |
+
"completions/max_length": 106.33333333333333,
|
| 579 |
+
"completions/max_terminated_length": 106.33333333333333,
|
| 580 |
+
"completions/mean_length": 60.489583333333336,
|
| 581 |
+
"completions/mean_terminated_length": 60.489583333333336,
|
| 582 |
+
"completions/min_length": 35.666666666666664,
|
| 583 |
+
"completions/min_terminated_length": 35.666666666666664,
|
| 584 |
+
"epoch": 0.0018649823803569398,
|
| 585 |
+
"frac_reward_zero_std": 0.5833333333333334,
|
| 586 |
+
"grad_norm": 0.0038077128119766712,
|
| 587 |
+
"kl": 0.0,
|
| 588 |
+
"learning_rate": 5.822472858062655e-07,
|
| 589 |
+
"loss": -0.00032259501516819,
|
| 590 |
+
"num_tokens": 1000028.0,
|
| 591 |
+
"reward": 9.934107462565104e-09,
|
| 592 |
+
"reward_std": 0.44259731968243915,
|
| 593 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 0.0,
|
| 594 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052845637004,
|
| 595 |
+
"step": 1050
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"clip_ratio/high_max": 0.0,
|
| 599 |
+
"clip_ratio/high_mean": 0.0,
|
| 600 |
+
"clip_ratio/low_mean": 6.211180239915847e-05,
|
| 601 |
+
"clip_ratio/low_min": 6.211180239915847e-05,
|
| 602 |
+
"clip_ratio/region_mean": 6.211180239915847e-05,
|
| 603 |
+
"completion_length": 62.45,
|
| 604 |
+
"completions/clipped_ratio": 0.0,
|
| 605 |
+
"completions/max_length": 102.66666666666667,
|
| 606 |
+
"completions/max_terminated_length": 102.66666666666667,
|
| 607 |
+
"completions/mean_length": 62.046875,
|
| 608 |
+
"completions/mean_terminated_length": 62.046875,
|
| 609 |
+
"completions/min_length": 38.0,
|
| 610 |
+
"completions/min_terminated_length": 38.0,
|
| 611 |
+
"epoch": 0.001953791065135842,
|
| 612 |
+
"frac_reward_zero_std": 0.5416666666666666,
|
| 613 |
+
"grad_norm": 0.010413075797259808,
|
| 614 |
+
"kl": 0.0,
|
| 615 |
+
"learning_rate": 6.099997779800627e-07,
|
| 616 |
+
"loss": 0.0010119295865297319,
|
| 617 |
+
"num_tokens": 1042829.0,
|
| 618 |
+
"reward": 1.241763432820638e-09,
|
| 619 |
+
"reward_std": 0.5440023342768351,
|
| 620 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.483526865641276e-09,
|
| 621 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 622 |
+
"step": 1100
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"clip_ratio/high_max": 0.0,
|
| 626 |
+
"clip_ratio/high_mean": 0.0,
|
| 627 |
+
"clip_ratio/low_mean": 5.3191487677395346e-05,
|
| 628 |
+
"clip_ratio/low_min": 5.3191487677395346e-05,
|
| 629 |
+
"clip_ratio/region_mean": 5.3191487677395346e-05,
|
| 630 |
+
"completion_length": 67.07,
|
| 631 |
+
"completions/clipped_ratio": 0.0,
|
| 632 |
+
"completions/max_length": 104.66666666666667,
|
| 633 |
+
"completions/max_terminated_length": 104.66666666666667,
|
| 634 |
+
"completions/mean_length": 67.171875,
|
| 635 |
+
"completions/mean_terminated_length": 67.171875,
|
| 636 |
+
"completions/min_length": 35.666666666666664,
|
| 637 |
+
"completions/min_terminated_length": 35.666666666666664,
|
| 638 |
+
"epoch": 0.0020425997499147436,
|
| 639 |
+
"frac_reward_zero_std": 0.4583333333333333,
|
| 640 |
+
"grad_norm": 0.0493728443980217,
|
| 641 |
+
"kl": 0.0,
|
| 642 |
+
"learning_rate": 6.377522701538599e-07,
|
| 643 |
+
"loss": -0.00021354854106903076,
|
| 644 |
+
"num_tokens": 1088622.0,
|
| 645 |
+
"reward": -1.7074247201283772e-08,
|
| 646 |
+
"reward_std": 0.5459979971249899,
|
| 647 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -2.0489096641540527e-08,
|
| 648 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052845637004,
|
| 649 |
+
"step": 1150
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"clip_ratio/high_max": 6.211180239915847e-05,
|
| 653 |
+
"clip_ratio/high_mean": 6.211180239915847e-05,
|
| 654 |
+
"clip_ratio/low_mean": 4.999999888241291e-05,
|
| 655 |
+
"clip_ratio/low_min": 4.999999888241291e-05,
|
| 656 |
+
"clip_ratio/region_mean": 0.00011211180128157139,
|
| 657 |
+
"completion_length": 66.76,
|
| 658 |
+
"completions/clipped_ratio": 0.0,
|
| 659 |
+
"completions/max_length": 100.33333333333333,
|
| 660 |
+
"completions/max_terminated_length": 100.33333333333333,
|
| 661 |
+
"completions/mean_length": 67.109375,
|
| 662 |
+
"completions/mean_terminated_length": 67.109375,
|
| 663 |
+
"completions/min_length": 40.666666666666664,
|
| 664 |
+
"completions/min_terminated_length": 40.666666666666664,
|
| 665 |
+
"epoch": 0.0021314084346936457,
|
| 666 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 667 |
+
"grad_norm": 0.0007921307114884257,
|
| 668 |
+
"kl": 0.0,
|
| 669 |
+
"learning_rate": 6.655047623276571e-07,
|
| 670 |
+
"loss": -0.0006299776583909988,
|
| 671 |
+
"num_tokens": 1134523.0,
|
| 672 |
+
"reward": 4.967053731282552e-09,
|
| 673 |
+
"reward_std": 0.24005796015262604,
|
| 674 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.4901161193847656e-08,
|
| 675 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 676 |
+
"step": 1200
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"clip_ratio/high_max": 0.0,
|
| 680 |
+
"clip_ratio/high_mean": 0.0,
|
| 681 |
+
"clip_ratio/low_mean": 0.0,
|
| 682 |
+
"clip_ratio/low_min": 0.0,
|
| 683 |
+
"clip_ratio/region_mean": 0.0,
|
| 684 |
+
"completion_length": 66.65,
|
| 685 |
+
"completions/clipped_ratio": 0.0,
|
| 686 |
+
"completions/max_length": 108.5,
|
| 687 |
+
"completions/max_terminated_length": 108.5,
|
| 688 |
+
"completions/mean_length": 64.26953125,
|
| 689 |
+
"completions/mean_terminated_length": 64.26953125,
|
| 690 |
+
"completions/min_length": 33.5,
|
| 691 |
+
"completions/min_terminated_length": 33.5,
|
| 692 |
+
"epoch": 0.0022202171194725473,
|
| 693 |
+
"frac_reward_zero_std": 0.6875,
|
| 694 |
+
"grad_norm": 0.0,
|
| 695 |
+
"kl": 0.0,
|
| 696 |
+
"learning_rate": 6.932572545014544e-07,
|
| 697 |
+
"loss": -0.000320572704076767,
|
| 698 |
+
"num_tokens": 1193424.0,
|
| 699 |
+
"reward": 3.725290298461914e-09,
|
| 700 |
+
"reward_std": 0.4126547574996948,
|
| 701 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.7939677238464355e-09,
|
| 702 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 703 |
+
"step": 1250
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"clip_ratio/high_max": 0.0,
|
| 707 |
+
"clip_ratio/high_mean": 0.0,
|
| 708 |
+
"clip_ratio/low_mean": 0.00013079616706818342,
|
| 709 |
+
"clip_ratio/low_min": 0.00013079616706818342,
|
| 710 |
+
"clip_ratio/region_mean": 0.00013079616706818342,
|
| 711 |
+
"completion_length": 65.935,
|
| 712 |
+
"completions/clipped_ratio": 0.0,
|
| 713 |
+
"completions/max_length": 100.33333333333333,
|
| 714 |
+
"completions/max_terminated_length": 100.33333333333333,
|
| 715 |
+
"completions/mean_length": 66.515625,
|
| 716 |
+
"completions/mean_terminated_length": 66.515625,
|
| 717 |
+
"completions/min_length": 35.333333333333336,
|
| 718 |
+
"completions/min_terminated_length": 35.333333333333336,
|
| 719 |
+
"epoch": 0.0023090258042514494,
|
| 720 |
+
"frac_reward_zero_std": 0.625,
|
| 721 |
+
"grad_norm": 0.009241115301847458,
|
| 722 |
+
"kl": 0.0,
|
| 723 |
+
"learning_rate": 7.210097466752515e-07,
|
| 724 |
+
"loss": -0.000618676021695137,
|
| 725 |
+
"num_tokens": 1239227.0,
|
| 726 |
+
"reward": 0.0,
|
| 727 |
+
"reward_std": 0.4066670338312785,
|
| 728 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 0.0,
|
| 729 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052845637004,
|
| 730 |
+
"step": 1300
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"clip_ratio/high_max": 0.0,
|
| 734 |
+
"clip_ratio/high_mean": 0.0,
|
| 735 |
+
"clip_ratio/low_mean": 5.586592014878988e-05,
|
| 736 |
+
"clip_ratio/low_min": 5.586592014878988e-05,
|
| 737 |
+
"clip_ratio/region_mean": 5.586592014878988e-05,
|
| 738 |
+
"completion_length": 61.28,
|
| 739 |
+
"completions/clipped_ratio": 0.0,
|
| 740 |
+
"completions/max_length": 106.66666666666667,
|
| 741 |
+
"completions/max_terminated_length": 106.66666666666667,
|
| 742 |
+
"completions/mean_length": 65.02083333333333,
|
| 743 |
+
"completions/mean_terminated_length": 65.02083333333333,
|
| 744 |
+
"completions/min_length": 33.666666666666664,
|
| 745 |
+
"completions/min_terminated_length": 33.666666666666664,
|
| 746 |
+
"epoch": 0.002397834489030351,
|
| 747 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 748 |
+
"grad_norm": 0.010475926101207733,
|
| 749 |
+
"kl": 0.0,
|
| 750 |
+
"learning_rate": 7.487622388490487e-07,
|
| 751 |
+
"loss": 0.0006281441450119019,
|
| 752 |
+
"num_tokens": 1283967.0,
|
| 753 |
+
"reward": 9.934107462565104e-09,
|
| 754 |
+
"reward_std": 0.3140339305003484,
|
| 755 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.241763432820638e-09,
|
| 756 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 757 |
+
"step": 1350
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"clip_ratio/high_max": 0.0,
|
| 761 |
+
"clip_ratio/high_mean": 0.0,
|
| 762 |
+
"clip_ratio/low_mean": 0.00010416666977107525,
|
| 763 |
+
"clip_ratio/low_min": 0.00010416666977107525,
|
| 764 |
+
"clip_ratio/region_mean": 0.00010416666977107525,
|
| 765 |
+
"completion_length": 64.255,
|
| 766 |
+
"completions/clipped_ratio": 0.0,
|
| 767 |
+
"completions/max_length": 96.66666666666667,
|
| 768 |
+
"completions/max_terminated_length": 96.66666666666667,
|
| 769 |
+
"completions/mean_length": 61.161458333333336,
|
| 770 |
+
"completions/mean_terminated_length": 61.161458333333336,
|
| 771 |
+
"completions/min_length": 34.333333333333336,
|
| 772 |
+
"completions/min_terminated_length": 34.333333333333336,
|
| 773 |
+
"epoch": 0.002486643173809253,
|
| 774 |
+
"frac_reward_zero_std": 0.7083333333333334,
|
| 775 |
+
"grad_norm": 0.02102189138531685,
|
| 776 |
+
"kl": 0.0,
|
| 777 |
+
"learning_rate": 7.765147310228459e-07,
|
| 778 |
+
"loss": -0.0013700899481773375,
|
| 779 |
+
"num_tokens": 1326542.0,
|
| 780 |
+
"reward": 2.483526865641276e-09,
|
| 781 |
+
"reward_std": 0.38102721174558,
|
| 782 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 7.450580596923828e-09,
|
| 783 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 784 |
+
"step": 1400
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"clip_ratio/high_max": 0.0,
|
| 788 |
+
"clip_ratio/high_mean": 0.0,
|
| 789 |
+
"clip_ratio/low_mean": 8.771929889917374e-05,
|
| 790 |
+
"clip_ratio/low_min": 8.771929889917374e-05,
|
| 791 |
+
"clip_ratio/region_mean": 8.771929889917374e-05,
|
| 792 |
+
"completion_length": 65.215,
|
| 793 |
+
"completions/clipped_ratio": 0.0,
|
| 794 |
+
"completions/max_length": 112.0,
|
| 795 |
+
"completions/max_terminated_length": 112.0,
|
| 796 |
+
"completions/mean_length": 67.92708333333333,
|
| 797 |
+
"completions/mean_terminated_length": 67.92708333333333,
|
| 798 |
+
"completions/min_length": 33.666666666666664,
|
| 799 |
+
"completions/min_terminated_length": 33.666666666666664,
|
| 800 |
+
"epoch": 0.002575451858588155,
|
| 801 |
+
"frac_reward_zero_std": 0.8333333333333334,
|
| 802 |
+
"grad_norm": 0.0,
|
| 803 |
+
"kl": 0.0,
|
| 804 |
+
"learning_rate": 8.042672231966431e-07,
|
| 805 |
+
"loss": 0.001429552286863327,
|
| 806 |
+
"num_tokens": 1373424.0,
|
| 807 |
+
"reward": 0.0,
|
| 808 |
+
"reward_std": 0.2193582976857821,
|
| 809 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 4.967053731282552e-09,
|
| 810 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 0.671936829884847,
|
| 811 |
+
"step": 1450
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"clip_ratio/high_max": 0.0,
|
| 815 |
+
"clip_ratio/high_mean": 0.0,
|
| 816 |
+
"clip_ratio/low_mean": 0.00015233600046485663,
|
| 817 |
+
"clip_ratio/low_min": 0.00015233600046485663,
|
| 818 |
+
"clip_ratio/region_mean": 0.00015233600046485663,
|
| 819 |
+
"completion_length": 69.035,
|
| 820 |
+
"completions/clipped_ratio": 0.0,
|
| 821 |
+
"completions/max_length": 94.0,
|
| 822 |
+
"completions/max_terminated_length": 94.0,
|
| 823 |
+
"completions/mean_length": 69.44791666666667,
|
| 824 |
+
"completions/mean_terminated_length": 69.44791666666667,
|
| 825 |
+
"completions/min_length": 41.0,
|
| 826 |
+
"completions/min_terminated_length": 41.0,
|
| 827 |
+
"epoch": 0.002664260543367057,
|
| 828 |
+
"frac_reward_zero_std": 0.625,
|
| 829 |
+
"grad_norm": 0.001688106800429523,
|
| 830 |
+
"kl": 0.0,
|
| 831 |
+
"learning_rate": 8.320197153704403e-07,
|
| 832 |
+
"loss": -0.0005409684777259826,
|
| 833 |
+
"num_tokens": 1421518.0,
|
| 834 |
+
"reward": 1.4901161193847656e-08,
|
| 835 |
+
"reward_std": 0.3605746428171794,
|
| 836 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.483526865641276e-08,
|
| 837 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 838 |
+
"step": 1500
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"eval_cer_subset": 0.012653280570618985,
|
| 842 |
+
"eval_cer_subset_edit_distance": 777,
|
| 843 |
+
"eval_cer_subset_groups": 250,
|
| 844 |
+
"eval_cer_subset_items": 250,
|
| 845 |
+
"eval_cer_subset_ref_chars": 61407,
|
| 846 |
+
"eval_loss": 0.03984795692112765,
|
| 847 |
+
"step": 1521
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"clip_ratio/high_max": 0.0,
|
| 851 |
+
"clip_ratio/high_mean": 0.0,
|
| 852 |
+
"clip_ratio/low_mean": 0.0002555130235850811,
|
| 853 |
+
"clip_ratio/low_min": 0.0002555130235850811,
|
| 854 |
+
"clip_ratio/region_mean": 0.0002555130235850811,
|
| 855 |
+
"completion_length": 68.995,
|
| 856 |
+
"completions/clipped_ratio": 0.0,
|
| 857 |
+
"completions/max_length": 104.66666666666667,
|
| 858 |
+
"completions/max_terminated_length": 104.66666666666667,
|
| 859 |
+
"completions/mean_length": 67.74479166666667,
|
| 860 |
+
"completions/mean_terminated_length": 67.74479166666667,
|
| 861 |
+
"completions/min_length": 38.666666666666664,
|
| 862 |
+
"completions/min_terminated_length": 38.666666666666664,
|
| 863 |
+
"epoch": 0.0027530692281459586,
|
| 864 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 865 |
+
"grad_norm": 0.01445477269589901,
|
| 866 |
+
"kl": 0.0,
|
| 867 |
+
"learning_rate": 8.597722075442375e-07,
|
| 868 |
+
"loss": -0.0012487749755382537,
|
| 869 |
+
"num_tokens": 1468197.0,
|
| 870 |
+
"reward": -1.241763432820638e-09,
|
| 871 |
+
"reward_std": 0.3732942243417104,
|
| 872 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -1.241763432820638e-09,
|
| 873 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 874 |
+
"step": 1550
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"clip_ratio/high_max": 0.0002725881431251764,
|
| 878 |
+
"clip_ratio/high_mean": 0.0002725881431251764,
|
| 879 |
+
"clip_ratio/low_mean": 0.00015432524960488082,
|
| 880 |
+
"clip_ratio/low_min": 0.00015432524960488082,
|
| 881 |
+
"clip_ratio/region_mean": 0.00042691339273005725,
|
| 882 |
+
"completion_length": 65.4,
|
| 883 |
+
"completions/clipped_ratio": 0.0,
|
| 884 |
+
"completions/max_length": 111.0,
|
| 885 |
+
"completions/max_terminated_length": 111.0,
|
| 886 |
+
"completions/mean_length": 65.203125,
|
| 887 |
+
"completions/mean_terminated_length": 65.203125,
|
| 888 |
+
"completions/min_length": 34.0,
|
| 889 |
+
"completions/min_terminated_length": 34.0,
|
| 890 |
+
"epoch": 0.0028418779129248607,
|
| 891 |
+
"frac_reward_zero_std": 0.5833333333333334,
|
| 892 |
+
"grad_norm": 0.0,
|
| 893 |
+
"kl": 0.0,
|
| 894 |
+
"learning_rate": 8.875246997180347e-07,
|
| 895 |
+
"loss": 0.0014201821386814117,
|
| 896 |
+
"num_tokens": 1513196.0,
|
| 897 |
+
"reward": 1.7384688059488933e-08,
|
| 898 |
+
"reward_std": 0.37156684199968976,
|
| 899 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 7.450580596923828e-09,
|
| 900 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 901 |
+
"step": 1600
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"clip_ratio/high_max": 0.0,
|
| 905 |
+
"clip_ratio/high_mean": 0.0,
|
| 906 |
+
"clip_ratio/low_mean": 7.8125e-05,
|
| 907 |
+
"clip_ratio/low_min": 7.8125e-05,
|
| 908 |
+
"clip_ratio/region_mean": 7.8125e-05,
|
| 909 |
+
"completion_length": 66.915,
|
| 910 |
+
"completions/clipped_ratio": 0.0,
|
| 911 |
+
"completions/max_length": 100.25,
|
| 912 |
+
"completions/max_terminated_length": 100.25,
|
| 913 |
+
"completions/mean_length": 67.140625,
|
| 914 |
+
"completions/mean_terminated_length": 67.140625,
|
| 915 |
+
"completions/min_length": 35.25,
|
| 916 |
+
"completions/min_terminated_length": 35.25,
|
| 917 |
+
"epoch": 0.002930686597703763,
|
| 918 |
+
"frac_reward_zero_std": 0.5,
|
| 919 |
+
"grad_norm": 0.00018169452960137278,
|
| 920 |
+
"kl": 0.0,
|
| 921 |
+
"learning_rate": 9.152771918918319e-07,
|
| 922 |
+
"loss": 0.0009048589318990707,
|
| 923 |
+
"num_tokens": 1573760.0,
|
| 924 |
+
"reward": 2.421438694000244e-08,
|
| 925 |
+
"reward_std": 0.35448721051216125,
|
| 926 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.2351741790771484e-08,
|
| 927 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 928 |
+
"step": 1650
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"clip_ratio/high_max": 0.00014303958974778652,
|
| 932 |
+
"clip_ratio/high_mean": 0.00014303958974778652,
|
| 933 |
+
"clip_ratio/low_mean": 6.75675692036748e-05,
|
| 934 |
+
"clip_ratio/low_min": 6.75675692036748e-05,
|
| 935 |
+
"clip_ratio/region_mean": 0.0002106071589514613,
|
| 936 |
+
"completion_length": 67.635,
|
| 937 |
+
"completions/clipped_ratio": 0.0,
|
| 938 |
+
"completions/max_length": 96.66666666666667,
|
| 939 |
+
"completions/max_terminated_length": 96.66666666666667,
|
| 940 |
+
"completions/mean_length": 64.96354166666667,
|
| 941 |
+
"completions/mean_terminated_length": 64.96354166666667,
|
| 942 |
+
"completions/min_length": 43.666666666666664,
|
| 943 |
+
"completions/min_terminated_length": 43.666666666666664,
|
| 944 |
+
"epoch": 0.0030194952824826645,
|
| 945 |
+
"frac_reward_zero_std": 0.7083333333333334,
|
| 946 |
+
"grad_norm": 0.09455703943967819,
|
| 947 |
+
"kl": 0.0,
|
| 948 |
+
"learning_rate": 9.430296840656291e-07,
|
| 949 |
+
"loss": -0.0004277878254652023,
|
| 950 |
+
"num_tokens": 1618329.0,
|
| 951 |
+
"reward": -9.934107462565104e-09,
|
| 952 |
+
"reward_std": 0.2702549050251643,
|
| 953 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -4.967053731282552e-09,
|
| 954 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052845637004,
|
| 955 |
+
"step": 1700
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"clip_ratio/high_max": 0.0,
|
| 959 |
+
"clip_ratio/high_mean": 0.0,
|
| 960 |
+
"clip_ratio/low_mean": 0.0,
|
| 961 |
+
"clip_ratio/low_min": 0.0,
|
| 962 |
+
"clip_ratio/region_mean": 0.0,
|
| 963 |
+
"completion_length": 59.285,
|
| 964 |
+
"completions/clipped_ratio": 0.0,
|
| 965 |
+
"completions/max_length": 106.66666666666667,
|
| 966 |
+
"completions/max_terminated_length": 106.66666666666667,
|
| 967 |
+
"completions/mean_length": 61.270833333333336,
|
| 968 |
+
"completions/mean_terminated_length": 61.270833333333336,
|
| 969 |
+
"completions/min_length": 27.666666666666668,
|
| 970 |
+
"completions/min_terminated_length": 27.666666666666668,
|
| 971 |
+
"epoch": 0.0031083039672615666,
|
| 972 |
+
"frac_reward_zero_std": 0.5833333333333334,
|
| 973 |
+
"grad_norm": 0.03458540141582489,
|
| 974 |
+
"kl": 0.0,
|
| 975 |
+
"learning_rate": 9.707821762394264e-07,
|
| 976 |
+
"loss": -0.0013288140296936035,
|
| 977 |
+
"num_tokens": 1660221.0,
|
| 978 |
+
"reward": -7.450580596923828e-09,
|
| 979 |
+
"reward_std": 0.3080942283074061,
|
| 980 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -2.483526865641276e-09,
|
| 981 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 982 |
+
"step": 1750
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"clip_ratio/high_max": 0.0,
|
| 986 |
+
"clip_ratio/high_mean": 0.0,
|
| 987 |
+
"clip_ratio/low_mean": 0.00017287629656493665,
|
| 988 |
+
"clip_ratio/low_min": 0.00017287629656493665,
|
| 989 |
+
"clip_ratio/region_mean": 0.00017287629656493665,
|
| 990 |
+
"completion_length": 68.225,
|
| 991 |
+
"completions/clipped_ratio": 0.0,
|
| 992 |
+
"completions/max_length": 101.0,
|
| 993 |
+
"completions/max_terminated_length": 101.0,
|
| 994 |
+
"completions/mean_length": 70.59895833333333,
|
| 995 |
+
"completions/mean_terminated_length": 70.59895833333333,
|
| 996 |
+
"completions/min_length": 45.333333333333336,
|
| 997 |
+
"completions/min_terminated_length": 45.333333333333336,
|
| 998 |
+
"epoch": 0.0031971126520404683,
|
| 999 |
+
"frac_reward_zero_std": 0.5833333333333334,
|
| 1000 |
+
"grad_norm": 0.011076130904257298,
|
| 1001 |
+
"kl": 0.0,
|
| 1002 |
+
"learning_rate": 9.985346684132236e-07,
|
| 1003 |
+
"loss": 0.00023731674998998643,
|
| 1004 |
+
"num_tokens": 1708216.0,
|
| 1005 |
+
"reward": 1.241763432820638e-08,
|
| 1006 |
+
"reward_std": 0.3638638307650884,
|
| 1007 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -7.450580596923828e-09,
|
| 1008 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052845637004,
|
| 1009 |
+
"step": 1800
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"clip_ratio/high_max": 0.00022903453558683395,
|
| 1013 |
+
"clip_ratio/high_mean": 0.00022903453558683395,
|
| 1014 |
+
"clip_ratio/low_mean": 0.00011979166883975267,
|
| 1015 |
+
"clip_ratio/low_min": 0.00011979166883975267,
|
| 1016 |
+
"clip_ratio/region_mean": 0.0003488262044265866,
|
| 1017 |
+
"completion_length": 70.16,
|
| 1018 |
+
"completions/clipped_ratio": 0.0,
|
| 1019 |
+
"completions/max_length": 120.33333333333333,
|
| 1020 |
+
"completions/max_terminated_length": 120.33333333333333,
|
| 1021 |
+
"completions/mean_length": 66.765625,
|
| 1022 |
+
"completions/mean_terminated_length": 66.765625,
|
| 1023 |
+
"completions/min_length": 39.333333333333336,
|
| 1024 |
+
"completions/min_terminated_length": 39.333333333333336,
|
| 1025 |
+
"epoch": 0.0032859213368193704,
|
| 1026 |
+
"frac_reward_zero_std": 0.5,
|
| 1027 |
+
"grad_norm": 0.03343971073627472,
|
| 1028 |
+
"kl": 0.0,
|
| 1029 |
+
"learning_rate": 1.0262871605870208e-06,
|
| 1030 |
+
"loss": -3.965016454458237e-05,
|
| 1031 |
+
"num_tokens": 1754323.0,
|
| 1032 |
+
"reward": 4.967053731282552e-09,
|
| 1033 |
+
"reward_std": 0.47326262791951496,
|
| 1034 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 4.967053731282552e-09,
|
| 1035 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1036 |
+
"step": 1850
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"clip_ratio/high_max": 0.0,
|
| 1040 |
+
"clip_ratio/high_mean": 0.0,
|
| 1041 |
+
"clip_ratio/low_mean": 0.00036457691341638563,
|
| 1042 |
+
"clip_ratio/low_min": 0.00036457691341638563,
|
| 1043 |
+
"clip_ratio/region_mean": 0.00036457691341638563,
|
| 1044 |
+
"completion_length": 64.08,
|
| 1045 |
+
"completions/clipped_ratio": 0.0,
|
| 1046 |
+
"completions/max_length": 98.0,
|
| 1047 |
+
"completions/max_terminated_length": 98.0,
|
| 1048 |
+
"completions/mean_length": 65.65625,
|
| 1049 |
+
"completions/mean_terminated_length": 65.65625,
|
| 1050 |
+
"completions/min_length": 31.333333333333332,
|
| 1051 |
+
"completions/min_terminated_length": 31.333333333333332,
|
| 1052 |
+
"epoch": 0.003374730021598272,
|
| 1053 |
+
"frac_reward_zero_std": 0.5833333333333334,
|
| 1054 |
+
"grad_norm": 0.03014628030359745,
|
| 1055 |
+
"kl": 0.0,
|
| 1056 |
+
"learning_rate": 1.054039652760818e-06,
|
| 1057 |
+
"loss": -0.0002728212811052799,
|
| 1058 |
+
"num_tokens": 1799249.0,
|
| 1059 |
+
"reward": 4.967053731282552e-09,
|
| 1060 |
+
"reward_std": 0.4208970566590627,
|
| 1061 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -2.483526865641276e-09,
|
| 1062 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1063 |
+
"step": 1900
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"clip_ratio/high_max": 0.0,
|
| 1067 |
+
"clip_ratio/high_mean": 0.0,
|
| 1068 |
+
"clip_ratio/low_mean": 8.547008968889713e-05,
|
| 1069 |
+
"clip_ratio/low_min": 8.547008968889713e-05,
|
| 1070 |
+
"clip_ratio/region_mean": 8.547008968889713e-05,
|
| 1071 |
+
"completion_length": 66.275,
|
| 1072 |
+
"completions/clipped_ratio": 0.0,
|
| 1073 |
+
"completions/max_length": 89.0,
|
| 1074 |
+
"completions/max_terminated_length": 89.0,
|
| 1075 |
+
"completions/mean_length": 65.98958333333333,
|
| 1076 |
+
"completions/mean_terminated_length": 65.98958333333333,
|
| 1077 |
+
"completions/min_length": 30.666666666666668,
|
| 1078 |
+
"completions/min_terminated_length": 30.666666666666668,
|
| 1079 |
+
"epoch": 0.003463538706377174,
|
| 1080 |
+
"frac_reward_zero_std": 0.5416666666666666,
|
| 1081 |
+
"grad_norm": 0.0011918310774490237,
|
| 1082 |
+
"kl": 0.0,
|
| 1083 |
+
"learning_rate": 1.0817921449346151e-06,
|
| 1084 |
+
"loss": -0.0011500194668769836,
|
| 1085 |
+
"num_tokens": 1844383.0,
|
| 1086 |
+
"reward": -7.450580596923828e-09,
|
| 1087 |
+
"reward_std": 0.4695793290932973,
|
| 1088 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.483526865641276e-09,
|
| 1089 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052845637004,
|
| 1090 |
+
"step": 1950
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"clip_ratio/high_max": 0.0001673903875052929,
|
| 1094 |
+
"clip_ratio/high_mean": 0.0001673903875052929,
|
| 1095 |
+
"clip_ratio/low_mean": 0.0002302072010934353,
|
| 1096 |
+
"clip_ratio/low_min": 0.0002302072010934353,
|
| 1097 |
+
"clip_ratio/region_mean": 0.00039759758859872816,
|
| 1098 |
+
"completion_length": 61.93,
|
| 1099 |
+
"completions/clipped_ratio": 0.0,
|
| 1100 |
+
"completions/max_length": 84.33333333333333,
|
| 1101 |
+
"completions/max_terminated_length": 84.33333333333333,
|
| 1102 |
+
"completions/mean_length": 61.59375,
|
| 1103 |
+
"completions/mean_terminated_length": 61.59375,
|
| 1104 |
+
"completions/min_length": 37.666666666666664,
|
| 1105 |
+
"completions/min_terminated_length": 37.666666666666664,
|
| 1106 |
+
"epoch": 0.003552347391156076,
|
| 1107 |
+
"frac_reward_zero_std": 0.5,
|
| 1108 |
+
"grad_norm": 0.02641136758029461,
|
| 1109 |
+
"kl": 0.0,
|
| 1110 |
+
"learning_rate": 1.1095446371084123e-06,
|
| 1111 |
+
"loss": 0.0008410254120826721,
|
| 1112 |
+
"num_tokens": 1886889.0,
|
| 1113 |
+
"reward": 1.4901161193847656e-08,
|
| 1114 |
+
"reward_std": 0.49693791071573895,
|
| 1115 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.4901161193847656e-08,
|
| 1116 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1117 |
+
"step": 2000
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"eval_cer_subset": 0.01279984366603156,
|
| 1121 |
+
"eval_cer_subset_edit_distance": 786,
|
| 1122 |
+
"eval_cer_subset_groups": 250,
|
| 1123 |
+
"eval_cer_subset_items": 250,
|
| 1124 |
+
"eval_cer_subset_ref_chars": 61407,
|
| 1125 |
+
"eval_loss": 0.04006968641114982,
|
| 1126 |
+
"step": 2028
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"clip_ratio/high_max": 0.0,
|
| 1130 |
+
"clip_ratio/high_mean": 0.0,
|
| 1131 |
+
"clip_ratio/low_mean": 9.708737954497337e-05,
|
| 1132 |
+
"clip_ratio/low_min": 9.708737954497337e-05,
|
| 1133 |
+
"clip_ratio/region_mean": 9.708737954497337e-05,
|
| 1134 |
+
"completion_length": 71.89,
|
| 1135 |
+
"completions/clipped_ratio": 0.0,
|
| 1136 |
+
"completions/max_length": 112.75,
|
| 1137 |
+
"completions/max_terminated_length": 112.75,
|
| 1138 |
+
"completions/mean_length": 69.125,
|
| 1139 |
+
"completions/mean_terminated_length": 69.125,
|
| 1140 |
+
"completions/min_length": 38.25,
|
| 1141 |
+
"completions/min_terminated_length": 38.25,
|
| 1142 |
+
"epoch": 0.003641156075934978,
|
| 1143 |
+
"frac_reward_zero_std": 0.53125,
|
| 1144 |
+
"grad_norm": 0.04017922654747963,
|
| 1145 |
+
"kl": 0.0,
|
| 1146 |
+
"learning_rate": 1.1372971292822095e-06,
|
| 1147 |
+
"loss": -1.7499616369605064e-05,
|
| 1148 |
+
"num_tokens": 1949777.0,
|
| 1149 |
+
"reward": 5.122274160385132e-09,
|
| 1150 |
+
"reward_std": 0.5594154596328735,
|
| 1151 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.3969838619232178e-09,
|
| 1152 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1153 |
+
"step": 2050
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"clip_ratio/high_max": 0.0,
|
| 1157 |
+
"clip_ratio/high_mean": 0.0,
|
| 1158 |
+
"clip_ratio/low_mean": 0.00032201658468693494,
|
| 1159 |
+
"clip_ratio/low_min": 0.00032201658468693494,
|
| 1160 |
+
"clip_ratio/region_mean": 0.00032201658468693494,
|
| 1161 |
+
"completion_length": 65.37,
|
| 1162 |
+
"completions/clipped_ratio": 0.0,
|
| 1163 |
+
"completions/max_length": 106.66666666666667,
|
| 1164 |
+
"completions/max_terminated_length": 106.66666666666667,
|
| 1165 |
+
"completions/mean_length": 67.13020833333333,
|
| 1166 |
+
"completions/mean_terminated_length": 67.13020833333333,
|
| 1167 |
+
"completions/min_length": 34.333333333333336,
|
| 1168 |
+
"completions/min_terminated_length": 34.333333333333336,
|
| 1169 |
+
"epoch": 0.0037299647607138796,
|
| 1170 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 1171 |
+
"grad_norm": 0.0,
|
| 1172 |
+
"kl": 0.0,
|
| 1173 |
+
"learning_rate": 1.1650496214560067e-06,
|
| 1174 |
+
"loss": -0.0009085966646671295,
|
| 1175 |
+
"num_tokens": 1995794.0,
|
| 1176 |
+
"reward": 1.241763432820638e-08,
|
| 1177 |
+
"reward_std": 0.3493447999159495,
|
| 1178 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.9868214925130207e-08,
|
| 1179 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1180 |
+
"step": 2100
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"clip_ratio/high_max": 0.0,
|
| 1184 |
+
"clip_ratio/high_mean": 0.0,
|
| 1185 |
+
"clip_ratio/low_mean": 5.8479532599449156e-05,
|
| 1186 |
+
"clip_ratio/low_min": 5.8479532599449156e-05,
|
| 1187 |
+
"clip_ratio/region_mean": 5.8479532599449156e-05,
|
| 1188 |
+
"completion_length": 69.465,
|
| 1189 |
+
"completions/clipped_ratio": 0.0,
|
| 1190 |
+
"completions/max_length": 102.33333333333333,
|
| 1191 |
+
"completions/max_terminated_length": 102.33333333333333,
|
| 1192 |
+
"completions/mean_length": 67.68229166666667,
|
| 1193 |
+
"completions/mean_terminated_length": 67.68229166666667,
|
| 1194 |
+
"completions/min_length": 41.666666666666664,
|
| 1195 |
+
"completions/min_terminated_length": 41.666666666666664,
|
| 1196 |
+
"epoch": 0.0038187734454927817,
|
| 1197 |
+
"frac_reward_zero_std": 0.7916666666666666,
|
| 1198 |
+
"grad_norm": 0.007103842683136463,
|
| 1199 |
+
"kl": 0.0,
|
| 1200 |
+
"learning_rate": 1.192802113629804e-06,
|
| 1201 |
+
"loss": -0.0002914236113429069,
|
| 1202 |
+
"num_tokens": 2042021.0,
|
| 1203 |
+
"reward": 0.0,
|
| 1204 |
+
"reward_std": 0.3860171635945638,
|
| 1205 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 0.0,
|
| 1206 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1207 |
+
"step": 2150
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"clip_ratio/high_max": 0.0,
|
| 1211 |
+
"clip_ratio/high_mean": 0.0,
|
| 1212 |
+
"clip_ratio/low_mean": 0.0,
|
| 1213 |
+
"clip_ratio/low_min": 0.0,
|
| 1214 |
+
"clip_ratio/region_mean": 0.0,
|
| 1215 |
+
"completion_length": 59.86,
|
| 1216 |
+
"completions/clipped_ratio": 0.0,
|
| 1217 |
+
"completions/max_length": 89.33333333333333,
|
| 1218 |
+
"completions/max_terminated_length": 89.33333333333333,
|
| 1219 |
+
"completions/mean_length": 58.645833333333336,
|
| 1220 |
+
"completions/mean_terminated_length": 58.645833333333336,
|
| 1221 |
+
"completions/min_length": 29.666666666666668,
|
| 1222 |
+
"completions/min_terminated_length": 29.666666666666668,
|
| 1223 |
+
"epoch": 0.003907582130271684,
|
| 1224 |
+
"frac_reward_zero_std": 0.7083333333333334,
|
| 1225 |
+
"grad_norm": 0.0,
|
| 1226 |
+
"kl": 0.0,
|
| 1227 |
+
"learning_rate": 1.220554605803601e-06,
|
| 1228 |
+
"loss": 0.0012667869031429291,
|
| 1229 |
+
"num_tokens": 2082697.0,
|
| 1230 |
+
"reward": 7.450580596923828e-09,
|
| 1231 |
+
"reward_std": 0.30209578076998395,
|
| 1232 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.241763432820638e-08,
|
| 1233 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1234 |
+
"step": 2200
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"clip_ratio/high_max": 0.00020011947490274906,
|
| 1238 |
+
"clip_ratio/high_mean": 0.00020011947490274906,
|
| 1239 |
+
"clip_ratio/low_mean": 0.00020966921467334033,
|
| 1240 |
+
"clip_ratio/low_min": 0.00020966921467334033,
|
| 1241 |
+
"clip_ratio/region_mean": 0.0004097886895760894,
|
| 1242 |
+
"completion_length": 64.93,
|
| 1243 |
+
"completions/clipped_ratio": 0.0,
|
| 1244 |
+
"completions/max_length": 107.0,
|
| 1245 |
+
"completions/max_terminated_length": 107.0,
|
| 1246 |
+
"completions/mean_length": 66.88020833333333,
|
| 1247 |
+
"completions/mean_terminated_length": 66.88020833333333,
|
| 1248 |
+
"completions/min_length": 34.333333333333336,
|
| 1249 |
+
"completions/min_terminated_length": 34.333333333333336,
|
| 1250 |
+
"epoch": 0.003996390815050585,
|
| 1251 |
+
"frac_reward_zero_std": 0.5833333333333334,
|
| 1252 |
+
"grad_norm": 0.0,
|
| 1253 |
+
"kl": 0.0,
|
| 1254 |
+
"learning_rate": 1.2483070979773983e-06,
|
| 1255 |
+
"loss": -0.0001166805624961853,
|
| 1256 |
+
"num_tokens": 2128290.0,
|
| 1257 |
+
"reward": 1.6142924626668293e-08,
|
| 1258 |
+
"reward_std": 0.4069610834121704,
|
| 1259 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 4.967053731282552e-09,
|
| 1260 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1261 |
+
"step": 2250
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"clip_ratio/high_max": 7.874015718698502e-05,
|
| 1265 |
+
"clip_ratio/high_mean": 7.874015718698502e-05,
|
| 1266 |
+
"clip_ratio/low_mean": 0.00014802721329033376,
|
| 1267 |
+
"clip_ratio/low_min": 0.00014802721329033376,
|
| 1268 |
+
"clip_ratio/region_mean": 0.00022676737047731875,
|
| 1269 |
+
"completion_length": 63.185,
|
| 1270 |
+
"completions/clipped_ratio": 0.0,
|
| 1271 |
+
"completions/max_length": 83.0,
|
| 1272 |
+
"completions/max_terminated_length": 83.0,
|
| 1273 |
+
"completions/mean_length": 64.86458333333333,
|
| 1274 |
+
"completions/mean_terminated_length": 64.86458333333333,
|
| 1275 |
+
"completions/min_length": 47.0,
|
| 1276 |
+
"completions/min_terminated_length": 47.0,
|
| 1277 |
+
"epoch": 0.004085199499829487,
|
| 1278 |
+
"frac_reward_zero_std": 0.625,
|
| 1279 |
+
"grad_norm": 0.007451566867530346,
|
| 1280 |
+
"kl": 0.0,
|
| 1281 |
+
"learning_rate": 1.2760595901511955e-06,
|
| 1282 |
+
"loss": -0.0005151259526610375,
|
| 1283 |
+
"num_tokens": 2172640.0,
|
| 1284 |
+
"reward": 6.20881716410319e-09,
|
| 1285 |
+
"reward_std": 0.48530252774556476,
|
| 1286 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 6.20881716410319e-09,
|
| 1287 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1288 |
+
"step": 2300
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"clip_ratio/high_max": 0.0,
|
| 1292 |
+
"clip_ratio/high_mean": 0.0,
|
| 1293 |
+
"clip_ratio/low_mean": 0.0003234845492988825,
|
| 1294 |
+
"clip_ratio/low_min": 0.0003234845492988825,
|
| 1295 |
+
"clip_ratio/region_mean": 0.0003234845492988825,
|
| 1296 |
+
"completion_length": 60.89,
|
| 1297 |
+
"completions/clipped_ratio": 0.0,
|
| 1298 |
+
"completions/max_length": 81.66666666666667,
|
| 1299 |
+
"completions/max_terminated_length": 81.66666666666667,
|
| 1300 |
+
"completions/mean_length": 59.354166666666664,
|
| 1301 |
+
"completions/mean_terminated_length": 59.354166666666664,
|
| 1302 |
+
"completions/min_length": 31.666666666666668,
|
| 1303 |
+
"completions/min_terminated_length": 31.666666666666668,
|
| 1304 |
+
"epoch": 0.004174008184608389,
|
| 1305 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 1306 |
+
"grad_norm": 0.00868611503392458,
|
| 1307 |
+
"kl": 0.0,
|
| 1308 |
+
"learning_rate": 1.3038120823249929e-06,
|
| 1309 |
+
"loss": -7.998313754796982e-05,
|
| 1310 |
+
"num_tokens": 2214372.0,
|
| 1311 |
+
"reward": 7.450580596923828e-09,
|
| 1312 |
+
"reward_std": 0.3333791295687358,
|
| 1313 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 7.450580596923828e-09,
|
| 1314 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1315 |
+
"step": 2350
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"clip_ratio/high_max": 0.0,
|
| 1319 |
+
"clip_ratio/high_mean": 0.0,
|
| 1320 |
+
"clip_ratio/low_mean": 6.451612804085016e-05,
|
| 1321 |
+
"clip_ratio/low_min": 6.451612804085016e-05,
|
| 1322 |
+
"clip_ratio/region_mean": 6.451612804085016e-05,
|
| 1323 |
+
"completion_length": 73.84,
|
| 1324 |
+
"completions/clipped_ratio": 0.0,
|
| 1325 |
+
"completions/max_length": 123.0,
|
| 1326 |
+
"completions/max_terminated_length": 123.0,
|
| 1327 |
+
"completions/mean_length": 74.765625,
|
| 1328 |
+
"completions/mean_terminated_length": 74.765625,
|
| 1329 |
+
"completions/min_length": 35.0,
|
| 1330 |
+
"completions/min_terminated_length": 35.0,
|
| 1331 |
+
"epoch": 0.004262816869387291,
|
| 1332 |
+
"frac_reward_zero_std": 0.7083333333333334,
|
| 1333 |
+
"grad_norm": 0.00028658212977461517,
|
| 1334 |
+
"kl": 0.0,
|
| 1335 |
+
"learning_rate": 1.33156457449879e-06,
|
| 1336 |
+
"loss": 0.00030371885746717454,
|
| 1337 |
+
"num_tokens": 2264791.0,
|
| 1338 |
+
"reward": 0.0,
|
| 1339 |
+
"reward_std": 0.22256243228912354,
|
| 1340 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 4.967053731282552e-09,
|
| 1341 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 0.671936829884847,
|
| 1342 |
+
"step": 2400
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"clip_ratio/high_max": 0.0002658057492226362,
|
| 1346 |
+
"clip_ratio/high_mean": 0.0002658057492226362,
|
| 1347 |
+
"clip_ratio/low_mean": 0.0006927049811929465,
|
| 1348 |
+
"clip_ratio/low_min": 0.0006927049811929465,
|
| 1349 |
+
"clip_ratio/region_mean": 0.0009585107304155827,
|
| 1350 |
+
"completion_length": 67.71,
|
| 1351 |
+
"completions/clipped_ratio": 0.0,
|
| 1352 |
+
"completions/max_length": 93.25,
|
| 1353 |
+
"completions/max_terminated_length": 93.25,
|
| 1354 |
+
"completions/mean_length": 67.19140625,
|
| 1355 |
+
"completions/mean_terminated_length": 67.19140625,
|
| 1356 |
+
"completions/min_length": 40.5,
|
| 1357 |
+
"completions/min_terminated_length": 40.5,
|
| 1358 |
+
"epoch": 0.004351625554166193,
|
| 1359 |
+
"frac_reward_zero_std": 0.46875,
|
| 1360 |
+
"grad_norm": 0.03887317702174187,
|
| 1361 |
+
"kl": 0.0,
|
| 1362 |
+
"learning_rate": 1.3593170666725873e-06,
|
| 1363 |
+
"loss": 0.0006680721044540405,
|
| 1364 |
+
"num_tokens": 2326168.0,
|
| 1365 |
+
"reward": -8.381903171539307e-09,
|
| 1366 |
+
"reward_std": 0.4901401996612549,
|
| 1367 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -3.725290298461914e-09,
|
| 1368 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1369 |
+
"step": 2450
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"clip_ratio/high_max": 0.0003085754346102476,
|
| 1373 |
+
"clip_ratio/high_mean": 0.0003085754346102476,
|
| 1374 |
+
"clip_ratio/low_mean": 0.00046078799292445185,
|
| 1375 |
+
"clip_ratio/low_min": 0.00046078799292445185,
|
| 1376 |
+
"clip_ratio/region_mean": 0.0007693634275346994,
|
| 1377 |
+
"completion_length": 63.165,
|
| 1378 |
+
"completions/clipped_ratio": 0.0,
|
| 1379 |
+
"completions/max_length": 92.0,
|
| 1380 |
+
"completions/max_terminated_length": 92.0,
|
| 1381 |
+
"completions/mean_length": 65.58333333333333,
|
| 1382 |
+
"completions/mean_terminated_length": 65.58333333333333,
|
| 1383 |
+
"completions/min_length": 44.0,
|
| 1384 |
+
"completions/min_terminated_length": 44.0,
|
| 1385 |
+
"epoch": 0.004440434238945095,
|
| 1386 |
+
"frac_reward_zero_std": 0.5416666666666666,
|
| 1387 |
+
"grad_norm": 0.009508333168923855,
|
| 1388 |
+
"kl": 0.0,
|
| 1389 |
+
"learning_rate": 1.3870695588463845e-06,
|
| 1390 |
+
"loss": -0.0017169666290283202,
|
| 1391 |
+
"num_tokens": 2371088.0,
|
| 1392 |
+
"reward": 2.483526865641276e-09,
|
| 1393 |
+
"reward_std": 0.46135319272677106,
|
| 1394 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -4.967053731282552e-09,
|
| 1395 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1396 |
+
"step": 2500
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"eval_cer_subset": 0.011969319458693634,
|
| 1400 |
+
"eval_cer_subset_edit_distance": 735,
|
| 1401 |
+
"eval_cer_subset_groups": 250,
|
| 1402 |
+
"eval_cer_subset_items": 250,
|
| 1403 |
+
"eval_cer_subset_ref_chars": 61407,
|
| 1404 |
+
"eval_loss": 0.04019638897687678,
|
| 1405 |
+
"step": 2535
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"clip_ratio/high_max": 0.0,
|
| 1409 |
+
"clip_ratio/high_mean": 0.0,
|
| 1410 |
+
"clip_ratio/low_mean": 0.0002695037797093391,
|
| 1411 |
+
"clip_ratio/low_min": 0.0002695037797093391,
|
| 1412 |
+
"clip_ratio/region_mean": 0.0002695037797093391,
|
| 1413 |
+
"completion_length": 70.675,
|
| 1414 |
+
"completions/clipped_ratio": 0.0,
|
| 1415 |
+
"completions/max_length": 120.33333333333333,
|
| 1416 |
+
"completions/max_terminated_length": 120.33333333333333,
|
| 1417 |
+
"completions/mean_length": 72.296875,
|
| 1418 |
+
"completions/mean_terminated_length": 72.296875,
|
| 1419 |
+
"completions/min_length": 35.333333333333336,
|
| 1420 |
+
"completions/min_terminated_length": 35.333333333333336,
|
| 1421 |
+
"epoch": 0.004529242923723997,
|
| 1422 |
+
"frac_reward_zero_std": 0.7083333333333334,
|
| 1423 |
+
"grad_norm": 0.008359396830201149,
|
| 1424 |
+
"kl": 0.0,
|
| 1425 |
+
"learning_rate": 1.4148220510201817e-06,
|
| 1426 |
+
"loss": 0.001508358269929886,
|
| 1427 |
+
"num_tokens": 2420241.0,
|
| 1428 |
+
"reward": -9.934107462565104e-09,
|
| 1429 |
+
"reward_std": 0.33431517084439594,
|
| 1430 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 0.0,
|
| 1431 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1432 |
+
"step": 2550
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"clip_ratio/high_max": 0.0,
|
| 1436 |
+
"clip_ratio/high_mean": 0.0,
|
| 1437 |
+
"clip_ratio/low_mean": 0.00021490163635462523,
|
| 1438 |
+
"clip_ratio/low_min": 0.00021490163635462523,
|
| 1439 |
+
"clip_ratio/region_mean": 0.00021490163635462523,
|
| 1440 |
+
"completion_length": 65.555,
|
| 1441 |
+
"completions/clipped_ratio": 0.0,
|
| 1442 |
+
"completions/max_length": 99.33333333333333,
|
| 1443 |
+
"completions/max_terminated_length": 99.33333333333333,
|
| 1444 |
+
"completions/mean_length": 60.026041666666664,
|
| 1445 |
+
"completions/mean_terminated_length": 60.026041666666664,
|
| 1446 |
+
"completions/min_length": 34.333333333333336,
|
| 1447 |
+
"completions/min_terminated_length": 34.333333333333336,
|
| 1448 |
+
"epoch": 0.004618051608502899,
|
| 1449 |
+
"frac_reward_zero_std": 0.5833333333333334,
|
| 1450 |
+
"grad_norm": 0.0035047000274062157,
|
| 1451 |
+
"kl": 0.0,
|
| 1452 |
+
"learning_rate": 1.4425745431939789e-06,
|
| 1453 |
+
"loss": -0.00014348618686199188,
|
| 1454 |
+
"num_tokens": 2461758.0,
|
| 1455 |
+
"reward": -7.450580596923828e-09,
|
| 1456 |
+
"reward_std": 0.3663780689239502,
|
| 1457 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.483526865641276e-09,
|
| 1458 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1459 |
+
"step": 2600
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"clip_ratio/high_max": 0.0,
|
| 1463 |
+
"clip_ratio/high_mean": 0.0,
|
| 1464 |
+
"clip_ratio/low_mean": 0.00012839030474424363,
|
| 1465 |
+
"clip_ratio/low_min": 0.00012839030474424363,
|
| 1466 |
+
"clip_ratio/region_mean": 0.00012839030474424363,
|
| 1467 |
+
"completion_length": 61.075,
|
| 1468 |
+
"completions/clipped_ratio": 0.0,
|
| 1469 |
+
"completions/max_length": 96.0,
|
| 1470 |
+
"completions/max_terminated_length": 96.0,
|
| 1471 |
+
"completions/mean_length": 60.666666666666664,
|
| 1472 |
+
"completions/mean_terminated_length": 60.666666666666664,
|
| 1473 |
+
"completions/min_length": 27.333333333333332,
|
| 1474 |
+
"completions/min_terminated_length": 27.333333333333332,
|
| 1475 |
+
"epoch": 0.004706860293281801,
|
| 1476 |
+
"frac_reward_zero_std": 0.5416666666666666,
|
| 1477 |
+
"grad_norm": 0.0020789902191609144,
|
| 1478 |
+
"kl": 0.0,
|
| 1479 |
+
"learning_rate": 1.470327035367776e-06,
|
| 1480 |
+
"loss": -0.0014517845213413239,
|
| 1481 |
+
"num_tokens": 2503798.0,
|
| 1482 |
+
"reward": 0.0,
|
| 1483 |
+
"reward_std": 0.287055641412735,
|
| 1484 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -4.967053731282552e-09,
|
| 1485 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1486 |
+
"step": 2650
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"clip_ratio/high_max": 4.999999888241291e-05,
|
| 1490 |
+
"clip_ratio/high_mean": 4.999999888241291e-05,
|
| 1491 |
+
"clip_ratio/low_mean": 0.0002898654108867049,
|
| 1492 |
+
"clip_ratio/low_min": 0.0002898654108867049,
|
| 1493 |
+
"clip_ratio/region_mean": 0.00033986540976911785,
|
| 1494 |
+
"completion_length": 57.85,
|
| 1495 |
+
"completions/clipped_ratio": 0.0,
|
| 1496 |
+
"completions/max_length": 95.33333333333333,
|
| 1497 |
+
"completions/max_terminated_length": 95.33333333333333,
|
| 1498 |
+
"completions/mean_length": 59.296875,
|
| 1499 |
+
"completions/mean_terminated_length": 59.296875,
|
| 1500 |
+
"completions/min_length": 23.666666666666668,
|
| 1501 |
+
"completions/min_terminated_length": 23.666666666666668,
|
| 1502 |
+
"epoch": 0.004795668978060702,
|
| 1503 |
+
"frac_reward_zero_std": 0.5416666666666666,
|
| 1504 |
+
"grad_norm": 0.0022281198762357235,
|
| 1505 |
+
"kl": 0.0,
|
| 1506 |
+
"learning_rate": 1.4980795275415732e-06,
|
| 1507 |
+
"loss": 0.0007117138057947159,
|
| 1508 |
+
"num_tokens": 2545047.0,
|
| 1509 |
+
"reward": 3.725290298461914e-09,
|
| 1510 |
+
"reward_std": 0.5628993709882101,
|
| 1511 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 3.725290298461914e-09,
|
| 1512 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1513 |
+
"step": 2700
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"clip_ratio/high_max": 0.0,
|
| 1517 |
+
"clip_ratio/high_mean": 0.0,
|
| 1518 |
+
"clip_ratio/low_mean": 0.00029827420599758624,
|
| 1519 |
+
"clip_ratio/low_min": 0.00029827420599758624,
|
| 1520 |
+
"clip_ratio/region_mean": 0.00029827420599758624,
|
| 1521 |
+
"completion_length": 69.345,
|
| 1522 |
+
"completions/clipped_ratio": 0.0,
|
| 1523 |
+
"completions/max_length": 107.33333333333333,
|
| 1524 |
+
"completions/max_terminated_length": 107.33333333333333,
|
| 1525 |
+
"completions/mean_length": 69.80208333333333,
|
| 1526 |
+
"completions/mean_terminated_length": 69.80208333333333,
|
| 1527 |
+
"completions/min_length": 35.666666666666664,
|
| 1528 |
+
"completions/min_terminated_length": 35.666666666666664,
|
| 1529 |
+
"epoch": 0.004884477662839604,
|
| 1530 |
+
"frac_reward_zero_std": 0.5416666666666666,
|
| 1531 |
+
"grad_norm": 0.0059377942234277725,
|
| 1532 |
+
"kl": 0.0,
|
| 1533 |
+
"learning_rate": 1.5258320197153704e-06,
|
| 1534 |
+
"loss": -0.000828000158071518,
|
| 1535 |
+
"num_tokens": 2592593.0,
|
| 1536 |
+
"reward": -2.483526865641276e-09,
|
| 1537 |
+
"reward_std": 0.4650723834832509,
|
| 1538 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.483526865641276e-09,
|
| 1539 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1540 |
+
"step": 2750
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"clip_ratio/high_max": 0.00023553025908768177,
|
| 1544 |
+
"clip_ratio/high_mean": 0.00023553025908768177,
|
| 1545 |
+
"clip_ratio/low_mean": 0.000865110270678997,
|
| 1546 |
+
"clip_ratio/low_min": 0.000865110270678997,
|
| 1547 |
+
"clip_ratio/region_mean": 0.0011006405297666788,
|
| 1548 |
+
"completion_length": 58.065,
|
| 1549 |
+
"completions/clipped_ratio": 0.0,
|
| 1550 |
+
"completions/max_length": 93.33333333333333,
|
| 1551 |
+
"completions/max_terminated_length": 93.33333333333333,
|
| 1552 |
+
"completions/mean_length": 57.572916666666664,
|
| 1553 |
+
"completions/mean_terminated_length": 57.572916666666664,
|
| 1554 |
+
"completions/min_length": 24.333333333333332,
|
| 1555 |
+
"completions/min_terminated_length": 24.333333333333332,
|
| 1556 |
+
"epoch": 0.004973286347618506,
|
| 1557 |
+
"frac_reward_zero_std": 0.4166666666666667,
|
| 1558 |
+
"grad_norm": 0.04159881919622421,
|
| 1559 |
+
"kl": 0.0,
|
| 1560 |
+
"learning_rate": 1.5535845118891678e-06,
|
| 1561 |
+
"loss": 0.0006424897909164429,
|
| 1562 |
+
"num_tokens": 2632703.0,
|
| 1563 |
+
"reward": -1.862645149230957e-09,
|
| 1564 |
+
"reward_std": 0.44419925411542255,
|
| 1565 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 3.725290298461914e-09,
|
| 1566 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1567 |
+
"step": 2800
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"clip_ratio/high_max": 0.00012865061871707438,
|
| 1571 |
+
"clip_ratio/high_mean": 0.00012865061871707438,
|
| 1572 |
+
"clip_ratio/low_mean": 0.000275228857062757,
|
| 1573 |
+
"clip_ratio/low_min": 0.000275228857062757,
|
| 1574 |
+
"clip_ratio/region_mean": 0.0004038794757798314,
|
| 1575 |
+
"completion_length": 63.82,
|
| 1576 |
+
"completions/clipped_ratio": 0.00390625,
|
| 1577 |
+
"completions/max_length": 168.25,
|
| 1578 |
+
"completions/max_terminated_length": 95.25,
|
| 1579 |
+
"completions/mean_length": 64.8203125,
|
| 1580 |
+
"completions/mean_terminated_length": 63.56938362121582,
|
| 1581 |
+
"completions/min_length": 37.25,
|
| 1582 |
+
"completions/min_terminated_length": 37.25,
|
| 1583 |
+
"epoch": 0.0050620950323974085,
|
| 1584 |
+
"frac_reward_zero_std": 0.59375,
|
| 1585 |
+
"grad_norm": 0.02093729004263878,
|
| 1586 |
+
"kl": 0.0,
|
| 1587 |
+
"learning_rate": 1.581337004062965e-06,
|
| 1588 |
+
"loss": -0.00023316640406847,
|
| 1589 |
+
"num_tokens": 2691505.0,
|
| 1590 |
+
"reward": -1.210719347000122e-08,
|
| 1591 |
+
"reward_std": 0.3830801360309124,
|
| 1592 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -1.1175870895385742e-08,
|
| 1593 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079053044319153,
|
| 1594 |
+
"step": 2850
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"clip_ratio/high_max": 0.0002845194097608328,
|
| 1598 |
+
"clip_ratio/high_mean": 0.0002845194097608328,
|
| 1599 |
+
"clip_ratio/low_mean": 2.2522523067891598e-05,
|
| 1600 |
+
"clip_ratio/low_min": 2.2522523067891598e-05,
|
| 1601 |
+
"clip_ratio/region_mean": 0.0003070419328287244,
|
| 1602 |
+
"completion_length": 62.65,
|
| 1603 |
+
"completions/clipped_ratio": 0.0,
|
| 1604 |
+
"completions/max_length": 89.33333333333333,
|
| 1605 |
+
"completions/max_terminated_length": 89.33333333333333,
|
| 1606 |
+
"completions/mean_length": 60.203125,
|
| 1607 |
+
"completions/mean_terminated_length": 60.203125,
|
| 1608 |
+
"completions/min_length": 32.666666666666664,
|
| 1609 |
+
"completions/min_terminated_length": 32.666666666666664,
|
| 1610 |
+
"epoch": 0.00515090371717631,
|
| 1611 |
+
"frac_reward_zero_std": 0.7083333333333334,
|
| 1612 |
+
"grad_norm": 0.005369533319026232,
|
| 1613 |
+
"kl": 0.0,
|
| 1614 |
+
"learning_rate": 1.6090894962367622e-06,
|
| 1615 |
+
"loss": 0.010556755065917968,
|
| 1616 |
+
"num_tokens": 2733288.0,
|
| 1617 |
+
"reward": 9.934107462565104e-09,
|
| 1618 |
+
"reward_std": 0.33344005048274994,
|
| 1619 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 9.934107462565104e-09,
|
| 1620 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1621 |
+
"step": 2900
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"clip_ratio/high_max": 0.0,
|
| 1625 |
+
"clip_ratio/high_mean": 0.0,
|
| 1626 |
+
"clip_ratio/low_mean": 0.0,
|
| 1627 |
+
"clip_ratio/low_min": 0.0,
|
| 1628 |
+
"clip_ratio/region_mean": 0.0,
|
| 1629 |
+
"completion_length": 69.76,
|
| 1630 |
+
"completions/clipped_ratio": 0.0,
|
| 1631 |
+
"completions/max_length": 109.33333333333333,
|
| 1632 |
+
"completions/max_terminated_length": 109.33333333333333,
|
| 1633 |
+
"completions/mean_length": 71.69270833333333,
|
| 1634 |
+
"completions/mean_terminated_length": 71.69270833333333,
|
| 1635 |
+
"completions/min_length": 41.0,
|
| 1636 |
+
"completions/min_terminated_length": 41.0,
|
| 1637 |
+
"epoch": 0.005239712401955212,
|
| 1638 |
+
"frac_reward_zero_std": 0.625,
|
| 1639 |
+
"grad_norm": 0.004602091386914253,
|
| 1640 |
+
"kl": 0.0,
|
| 1641 |
+
"learning_rate": 1.6368419884105594e-06,
|
| 1642 |
+
"loss": 0.00016404924914240838,
|
| 1643 |
+
"num_tokens": 2782853.0,
|
| 1644 |
+
"reward": -7.450580596923828e-09,
|
| 1645 |
+
"reward_std": 0.3352220952510834,
|
| 1646 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -1.241763432820638e-08,
|
| 1647 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1648 |
+
"step": 2950
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"clip_ratio/high_max": 0.00036425903439521787,
|
| 1652 |
+
"clip_ratio/high_mean": 0.00036425903439521787,
|
| 1653 |
+
"clip_ratio/low_mean": 0.0009387264680117369,
|
| 1654 |
+
"clip_ratio/low_min": 0.0009387264680117369,
|
| 1655 |
+
"clip_ratio/region_mean": 0.0013029855024069548,
|
| 1656 |
+
"completion_length": 59.11,
|
| 1657 |
+
"completions/clipped_ratio": 0.0,
|
| 1658 |
+
"completions/max_length": 90.33333333333333,
|
| 1659 |
+
"completions/max_terminated_length": 90.33333333333333,
|
| 1660 |
+
"completions/mean_length": 58.833333333333336,
|
| 1661 |
+
"completions/mean_terminated_length": 58.833333333333336,
|
| 1662 |
+
"completions/min_length": 29.666666666666668,
|
| 1663 |
+
"completions/min_terminated_length": 29.666666666666668,
|
| 1664 |
+
"epoch": 0.005328521086734114,
|
| 1665 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 1666 |
+
"grad_norm": 0.0,
|
| 1667 |
+
"kl": 0.0,
|
| 1668 |
+
"learning_rate": 1.6645944805843566e-06,
|
| 1669 |
+
"loss": -0.0011630910634994507,
|
| 1670 |
+
"num_tokens": 2823621.0,
|
| 1671 |
+
"reward": 2.483526865641276e-09,
|
| 1672 |
+
"reward_std": 0.4679803748925527,
|
| 1673 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.483526865641276e-09,
|
| 1674 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1675 |
+
"step": 3000
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"eval_cer_subset": 0.01195303467031446,
|
| 1679 |
+
"eval_cer_subset_edit_distance": 734,
|
| 1680 |
+
"eval_cer_subset_groups": 250,
|
| 1681 |
+
"eval_cer_subset_items": 250,
|
| 1682 |
+
"eval_cer_subset_ref_chars": 61407,
|
| 1683 |
+
"eval_loss": 0.04089325308837504,
|
| 1684 |
+
"step": 3042
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"clip_ratio/high_max": 0.0,
|
| 1688 |
+
"clip_ratio/high_mean": 0.0,
|
| 1689 |
+
"clip_ratio/low_mean": 0.00026520540937781335,
|
| 1690 |
+
"clip_ratio/low_min": 0.00026520540937781335,
|
| 1691 |
+
"clip_ratio/region_mean": 0.00026520540937781335,
|
| 1692 |
+
"completion_length": 71.79,
|
| 1693 |
+
"completions/clipped_ratio": 0.0,
|
| 1694 |
+
"completions/max_length": 105.33333333333333,
|
| 1695 |
+
"completions/max_terminated_length": 105.33333333333333,
|
| 1696 |
+
"completions/mean_length": 70.38541666666667,
|
| 1697 |
+
"completions/mean_terminated_length": 70.38541666666667,
|
| 1698 |
+
"completions/min_length": 37.0,
|
| 1699 |
+
"completions/min_terminated_length": 37.0,
|
| 1700 |
+
"epoch": 0.005417329771513016,
|
| 1701 |
+
"frac_reward_zero_std": 0.7083333333333334,
|
| 1702 |
+
"grad_norm": 0.012031897902488708,
|
| 1703 |
+
"kl": 0.0,
|
| 1704 |
+
"learning_rate": 1.6923469727581538e-06,
|
| 1705 |
+
"loss": 0.0010356108099222182,
|
| 1706 |
+
"num_tokens": 2872095.0,
|
| 1707 |
+
"reward": -3.725290298461914e-09,
|
| 1708 |
+
"reward_std": 0.40887144207954407,
|
| 1709 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 3.725290298461914e-09,
|
| 1710 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1711 |
+
"step": 3050
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"clip_ratio/high_max": 0.0,
|
| 1715 |
+
"clip_ratio/high_mean": 0.0,
|
| 1716 |
+
"clip_ratio/low_mean": 0.00041877904906868936,
|
| 1717 |
+
"clip_ratio/low_min": 0.00041877904906868936,
|
| 1718 |
+
"clip_ratio/region_mean": 0.00041877904906868936,
|
| 1719 |
+
"completion_length": 73.095,
|
| 1720 |
+
"completions/clipped_ratio": 0.0,
|
| 1721 |
+
"completions/max_length": 124.0,
|
| 1722 |
+
"completions/max_terminated_length": 124.0,
|
| 1723 |
+
"completions/mean_length": 75.04166666666667,
|
| 1724 |
+
"completions/mean_terminated_length": 75.04166666666667,
|
| 1725 |
+
"completions/min_length": 31.666666666666668,
|
| 1726 |
+
"completions/min_terminated_length": 31.666666666666668,
|
| 1727 |
+
"epoch": 0.005506138456291917,
|
| 1728 |
+
"frac_reward_zero_std": 0.5416666666666666,
|
| 1729 |
+
"grad_norm": 0.0002992149966303259,
|
| 1730 |
+
"kl": 0.0,
|
| 1731 |
+
"learning_rate": 1.720099464931951e-06,
|
| 1732 |
+
"loss": 0.0014635683596134186,
|
| 1733 |
+
"num_tokens": 2923015.0,
|
| 1734 |
+
"reward": -1.241763432820638e-08,
|
| 1735 |
+
"reward_std": 0.40434783697128296,
|
| 1736 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.483526865641276e-09,
|
| 1737 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1738 |
+
"step": 3100
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"clip_ratio/high_max": 0.0,
|
| 1742 |
+
"clip_ratio/high_mean": 0.0,
|
| 1743 |
+
"clip_ratio/low_mean": 0.0002545050159096718,
|
| 1744 |
+
"clip_ratio/low_min": 0.0002545050159096718,
|
| 1745 |
+
"clip_ratio/region_mean": 0.0002545050159096718,
|
| 1746 |
+
"completion_length": 65.495,
|
| 1747 |
+
"completions/clipped_ratio": 0.0,
|
| 1748 |
+
"completions/max_length": 101.33333333333333,
|
| 1749 |
+
"completions/max_terminated_length": 101.33333333333333,
|
| 1750 |
+
"completions/mean_length": 64.59895833333333,
|
| 1751 |
+
"completions/mean_terminated_length": 64.59895833333333,
|
| 1752 |
+
"completions/min_length": 38.0,
|
| 1753 |
+
"completions/min_terminated_length": 38.0,
|
| 1754 |
+
"epoch": 0.005594947141070819,
|
| 1755 |
+
"frac_reward_zero_std": 0.375,
|
| 1756 |
+
"grad_norm": 0.004821412730962038,
|
| 1757 |
+
"kl": 0.0,
|
| 1758 |
+
"learning_rate": 1.7478519571057482e-06,
|
| 1759 |
+
"loss": -0.0013929881155490876,
|
| 1760 |
+
"num_tokens": 2967578.0,
|
| 1761 |
+
"reward": 3.725290298461914e-09,
|
| 1762 |
+
"reward_std": 0.47509239117304486,
|
| 1763 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 0.0,
|
| 1764 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1765 |
+
"step": 3150
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"clip_ratio/high_max": 0.0,
|
| 1769 |
+
"clip_ratio/high_mean": 0.0,
|
| 1770 |
+
"clip_ratio/low_mean": 0.0,
|
| 1771 |
+
"clip_ratio/low_min": 0.0,
|
| 1772 |
+
"clip_ratio/region_mean": 0.0,
|
| 1773 |
+
"completion_length": 65.295,
|
| 1774 |
+
"completions/clipped_ratio": 0.0,
|
| 1775 |
+
"completions/max_length": 98.66666666666667,
|
| 1776 |
+
"completions/max_terminated_length": 98.66666666666667,
|
| 1777 |
+
"completions/mean_length": 65.95833333333333,
|
| 1778 |
+
"completions/mean_terminated_length": 65.95833333333333,
|
| 1779 |
+
"completions/min_length": 41.666666666666664,
|
| 1780 |
+
"completions/min_terminated_length": 41.666666666666664,
|
| 1781 |
+
"epoch": 0.0056837558258497215,
|
| 1782 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 1783 |
+
"grad_norm": 0.012544830329716206,
|
| 1784 |
+
"kl": 0.0,
|
| 1785 |
+
"learning_rate": 1.7756044492795454e-06,
|
| 1786 |
+
"loss": 0.0002492551319301128,
|
| 1787 |
+
"num_tokens": 3012562.0,
|
| 1788 |
+
"reward": -1.4901161193847656e-08,
|
| 1789 |
+
"reward_std": 0.31818433851003647,
|
| 1790 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -1.241763432820638e-08,
|
| 1791 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052845637004,
|
| 1792 |
+
"step": 3200
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"clip_ratio/high_max": 0.00016690524760633706,
|
| 1796 |
+
"clip_ratio/high_mean": 0.00016690524760633706,
|
| 1797 |
+
"clip_ratio/low_mean": 0.00013946280814707278,
|
| 1798 |
+
"clip_ratio/low_min": 0.00013946280814707278,
|
| 1799 |
+
"clip_ratio/region_mean": 0.00030636805575340987,
|
| 1800 |
+
"completion_length": 71.945,
|
| 1801 |
+
"completions/clipped_ratio": 0.0,
|
| 1802 |
+
"completions/max_length": 102.0,
|
| 1803 |
+
"completions/max_terminated_length": 102.0,
|
| 1804 |
+
"completions/mean_length": 70.26171875,
|
| 1805 |
+
"completions/mean_terminated_length": 70.26171875,
|
| 1806 |
+
"completions/min_length": 39.0,
|
| 1807 |
+
"completions/min_terminated_length": 39.0,
|
| 1808 |
+
"epoch": 0.005772564510628624,
|
| 1809 |
+
"frac_reward_zero_std": 0.625,
|
| 1810 |
+
"grad_norm": 0.004245330113917589,
|
| 1811 |
+
"kl": 0.0,
|
| 1812 |
+
"learning_rate": 1.8033569414533426e-06,
|
| 1813 |
+
"loss": 0.0010712506622076034,
|
| 1814 |
+
"num_tokens": 3076509.0,
|
| 1815 |
+
"reward": 3.725290298461914e-09,
|
| 1816 |
+
"reward_std": 0.46907638758420944,
|
| 1817 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 3.725290298461914e-09,
|
| 1818 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.007905274629593,
|
| 1819 |
+
"step": 3250
|
| 1820 |
+
},
|
| 1821 |
+
{
|
| 1822 |
+
"clip_ratio/high_max": 0.00015352941583842038,
|
| 1823 |
+
"clip_ratio/high_mean": 0.00015352941583842038,
|
| 1824 |
+
"clip_ratio/low_mean": 0.00021474448498338462,
|
| 1825 |
+
"clip_ratio/low_min": 0.00021474448498338462,
|
| 1826 |
+
"clip_ratio/region_mean": 0.000368273900821805,
|
| 1827 |
+
"completion_length": 67.95,
|
| 1828 |
+
"completions/clipped_ratio": 0.0,
|
| 1829 |
+
"completions/max_length": 114.66666666666667,
|
| 1830 |
+
"completions/max_terminated_length": 114.66666666666667,
|
| 1831 |
+
"completions/mean_length": 66.70833333333333,
|
| 1832 |
+
"completions/mean_terminated_length": 66.70833333333333,
|
| 1833 |
+
"completions/min_length": 28.666666666666668,
|
| 1834 |
+
"completions/min_terminated_length": 28.666666666666668,
|
| 1835 |
+
"epoch": 0.005861373195407526,
|
| 1836 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 1837 |
+
"grad_norm": 0.0,
|
| 1838 |
+
"kl": 0.0,
|
| 1839 |
+
"learning_rate": 1.8311094336271398e-06,
|
| 1840 |
+
"loss": -0.0011709482222795487,
|
| 1841 |
+
"num_tokens": 3121613.0,
|
| 1842 |
+
"reward": 0.0,
|
| 1843 |
+
"reward_std": 0.3044033149878184,
|
| 1844 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 0.0,
|
| 1845 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1846 |
+
"step": 3300
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"clip_ratio/high_max": 0.000174296242184937,
|
| 1850 |
+
"clip_ratio/high_mean": 0.000174296242184937,
|
| 1851 |
+
"clip_ratio/low_mean": 0.0004580108635127544,
|
| 1852 |
+
"clip_ratio/low_min": 0.0004580108635127544,
|
| 1853 |
+
"clip_ratio/region_mean": 0.0006323071056976914,
|
| 1854 |
+
"completion_length": 62.405,
|
| 1855 |
+
"completions/clipped_ratio": 0.0,
|
| 1856 |
+
"completions/max_length": 111.66666666666667,
|
| 1857 |
+
"completions/max_terminated_length": 111.66666666666667,
|
| 1858 |
+
"completions/mean_length": 63.166666666666664,
|
| 1859 |
+
"completions/mean_terminated_length": 63.166666666666664,
|
| 1860 |
+
"completions/min_length": 34.666666666666664,
|
| 1861 |
+
"completions/min_terminated_length": 34.666666666666664,
|
| 1862 |
+
"epoch": 0.005950181880186427,
|
| 1863 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 1864 |
+
"grad_norm": 0.004354190081357956,
|
| 1865 |
+
"kl": 0.0,
|
| 1866 |
+
"learning_rate": 1.858861925800937e-06,
|
| 1867 |
+
"loss": 0.001297842264175415,
|
| 1868 |
+
"num_tokens": 3165293.0,
|
| 1869 |
+
"reward": -4.967053731282552e-09,
|
| 1870 |
+
"reward_std": 0.3569997350374858,
|
| 1871 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -2.483526865641276e-09,
|
| 1872 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1873 |
+
"step": 3350
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"clip_ratio/high_max": 0.0,
|
| 1877 |
+
"clip_ratio/high_mean": 0.0,
|
| 1878 |
+
"clip_ratio/low_mean": 0.0005731566715985537,
|
| 1879 |
+
"clip_ratio/low_min": 0.0005731566715985537,
|
| 1880 |
+
"clip_ratio/region_mean": 0.0005731566715985537,
|
| 1881 |
+
"completion_length": 54.945,
|
| 1882 |
+
"completions/clipped_ratio": 0.0,
|
| 1883 |
+
"completions/max_length": 82.0,
|
| 1884 |
+
"completions/max_terminated_length": 82.0,
|
| 1885 |
+
"completions/mean_length": 53.213541666666664,
|
| 1886 |
+
"completions/mean_terminated_length": 53.213541666666664,
|
| 1887 |
+
"completions/min_length": 26.0,
|
| 1888 |
+
"completions/min_terminated_length": 26.0,
|
| 1889 |
+
"epoch": 0.006038990564965329,
|
| 1890 |
+
"frac_reward_zero_std": 0.5,
|
| 1891 |
+
"grad_norm": 0.031956613063812256,
|
| 1892 |
+
"kl": 0.0,
|
| 1893 |
+
"learning_rate": 1.8866144179747341e-06,
|
| 1894 |
+
"loss": -0.0013629911839962005,
|
| 1895 |
+
"num_tokens": 3202198.0,
|
| 1896 |
+
"reward": 1.0554989178975424e-08,
|
| 1897 |
+
"reward_std": 0.40813971559206647,
|
| 1898 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.0554989178975424e-08,
|
| 1899 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1900 |
+
"step": 3400
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"clip_ratio/high_max": 0.00011518601793795824,
|
| 1904 |
+
"clip_ratio/high_mean": 0.00011518601793795824,
|
| 1905 |
+
"clip_ratio/low_mean": 0.00020821858197450638,
|
| 1906 |
+
"clip_ratio/low_min": 0.00020821858197450638,
|
| 1907 |
+
"clip_ratio/region_mean": 0.0003234045999124646,
|
| 1908 |
+
"completion_length": 67.62,
|
| 1909 |
+
"completions/clipped_ratio": 0.0,
|
| 1910 |
+
"completions/max_length": 119.33333333333333,
|
| 1911 |
+
"completions/max_terminated_length": 119.33333333333333,
|
| 1912 |
+
"completions/mean_length": 70.0625,
|
| 1913 |
+
"completions/mean_terminated_length": 70.0625,
|
| 1914 |
+
"completions/min_length": 43.333333333333336,
|
| 1915 |
+
"completions/min_terminated_length": 43.333333333333336,
|
| 1916 |
+
"epoch": 0.006127799249744231,
|
| 1917 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 1918 |
+
"grad_norm": 0.0,
|
| 1919 |
+
"kl": 0.0,
|
| 1920 |
+
"learning_rate": 1.9143669101485313e-06,
|
| 1921 |
+
"loss": -0.0013814741373062134,
|
| 1922 |
+
"num_tokens": 3250042.0,
|
| 1923 |
+
"reward": 0.0,
|
| 1924 |
+
"reward_std": 0.415747731924057,
|
| 1925 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 0.0,
|
| 1926 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1927 |
+
"step": 3450
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"clip_ratio/high_max": 0.00010309278033673763,
|
| 1931 |
+
"clip_ratio/high_mean": 0.00010309278033673763,
|
| 1932 |
+
"clip_ratio/low_mean": 0.0007092890702188015,
|
| 1933 |
+
"clip_ratio/low_min": 0.0007092890702188015,
|
| 1934 |
+
"clip_ratio/region_mean": 0.0008123818505555391,
|
| 1935 |
+
"completion_length": 67.305,
|
| 1936 |
+
"completions/clipped_ratio": 0.0,
|
| 1937 |
+
"completions/max_length": 107.66666666666667,
|
| 1938 |
+
"completions/max_terminated_length": 107.66666666666667,
|
| 1939 |
+
"completions/mean_length": 66.76041666666667,
|
| 1940 |
+
"completions/mean_terminated_length": 66.76041666666667,
|
| 1941 |
+
"completions/min_length": 37.666666666666664,
|
| 1942 |
+
"completions/min_terminated_length": 37.666666666666664,
|
| 1943 |
+
"epoch": 0.006216607934523133,
|
| 1944 |
+
"frac_reward_zero_std": 0.6666666666666666,
|
| 1945 |
+
"grad_norm": 0.0,
|
| 1946 |
+
"kl": 0.0,
|
| 1947 |
+
"learning_rate": 1.942119402322329e-06,
|
| 1948 |
+
"loss": 0.0015093806385993959,
|
| 1949 |
+
"num_tokens": 3295476.0,
|
| 1950 |
+
"reward": -7.450580596923828e-09,
|
| 1951 |
+
"reward_std": 0.32829682032267254,
|
| 1952 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -2.483526865641276e-09,
|
| 1953 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1954 |
+
"step": 3500
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"eval_cer_subset": 0.01231130001465631,
|
| 1958 |
+
"eval_cer_subset_edit_distance": 756,
|
| 1959 |
+
"eval_cer_subset_groups": 250,
|
| 1960 |
+
"eval_cer_subset_items": 250,
|
| 1961 |
+
"eval_cer_subset_ref_chars": 61407,
|
| 1962 |
+
"eval_loss": 0.041400063351282865,
|
| 1963 |
+
"step": 3549
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"clip_ratio/high_max": 0.0,
|
| 1967 |
+
"clip_ratio/high_mean": 0.0,
|
| 1968 |
+
"clip_ratio/low_mean": 6.250000093132258e-05,
|
| 1969 |
+
"clip_ratio/low_min": 6.250000093132258e-05,
|
| 1970 |
+
"clip_ratio/region_mean": 6.250000093132258e-05,
|
| 1971 |
+
"completion_length": 68.47,
|
| 1972 |
+
"completions/clipped_ratio": 0.0,
|
| 1973 |
+
"completions/max_length": 113.0,
|
| 1974 |
+
"completions/max_terminated_length": 113.0,
|
| 1975 |
+
"completions/mean_length": 69.28645833333333,
|
| 1976 |
+
"completions/mean_terminated_length": 69.28645833333333,
|
| 1977 |
+
"completions/min_length": 37.666666666666664,
|
| 1978 |
+
"completions/min_terminated_length": 37.666666666666664,
|
| 1979 |
+
"epoch": 0.0063054166193020345,
|
| 1980 |
+
"frac_reward_zero_std": 0.375,
|
| 1981 |
+
"grad_norm": 0.015265848487615585,
|
| 1982 |
+
"kl": 0.0,
|
| 1983 |
+
"learning_rate": 1.969871894496126e-06,
|
| 1984 |
+
"loss": 0.0004761111736297607,
|
| 1985 |
+
"num_tokens": 3342339.0,
|
| 1986 |
+
"reward": 1.1175870895385742e-08,
|
| 1987 |
+
"reward_std": 0.5635309418042501,
|
| 1988 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.6142924626668293e-08,
|
| 1989 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 1990 |
+
"step": 3550
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"clip_ratio/high_max": 0.0002064430946484208,
|
| 1994 |
+
"clip_ratio/high_mean": 0.0002064430946484208,
|
| 1995 |
+
"clip_ratio/low_mean": 0.0006543642934411765,
|
| 1996 |
+
"clip_ratio/low_min": 0.0006543642934411765,
|
| 1997 |
+
"clip_ratio/region_mean": 0.0008608073880895972,
|
| 1998 |
+
"completion_length": 62.445,
|
| 1999 |
+
"completions/clipped_ratio": 0.0,
|
| 2000 |
+
"completions/max_length": 95.66666666666667,
|
| 2001 |
+
"completions/max_terminated_length": 95.66666666666667,
|
| 2002 |
+
"completions/mean_length": 62.0,
|
| 2003 |
+
"completions/mean_terminated_length": 62.0,
|
| 2004 |
+
"completions/min_length": 39.666666666666664,
|
| 2005 |
+
"completions/min_terminated_length": 39.666666666666664,
|
| 2006 |
+
"epoch": 0.0063942253040809366,
|
| 2007 |
+
"frac_reward_zero_std": 0.5,
|
| 2008 |
+
"grad_norm": 0.012917496263980865,
|
| 2009 |
+
"kl": 0.0,
|
| 2010 |
+
"learning_rate": 1.9976243866699233e-06,
|
| 2011 |
+
"loss": -0.0004654873907566071,
|
| 2012 |
+
"num_tokens": 3385203.0,
|
| 2013 |
+
"reward": -7.450580596923828e-09,
|
| 2014 |
+
"reward_std": 0.3997733195622762,
|
| 2015 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -2.483526865641276e-09,
|
| 2016 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 2017 |
+
"step": 3600
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"clip_ratio/high_max": 0.00014757968951016663,
|
| 2021 |
+
"clip_ratio/high_mean": 0.00014757968951016663,
|
| 2022 |
+
"clip_ratio/low_mean": 0.00023364254739135503,
|
| 2023 |
+
"clip_ratio/low_min": 0.00023364254739135503,
|
| 2024 |
+
"clip_ratio/region_mean": 0.0003812222369015217,
|
| 2025 |
+
"completion_length": 56.87,
|
| 2026 |
+
"completions/clipped_ratio": 0.0,
|
| 2027 |
+
"completions/max_length": 86.25,
|
| 2028 |
+
"completions/max_terminated_length": 86.25,
|
| 2029 |
+
"completions/mean_length": 57.1484375,
|
| 2030 |
+
"completions/mean_terminated_length": 57.1484375,
|
| 2031 |
+
"completions/min_length": 30.5,
|
| 2032 |
+
"completions/min_terminated_length": 30.5,
|
| 2033 |
+
"epoch": 0.006483033988859839,
|
| 2034 |
+
"frac_reward_zero_std": 0.625,
|
| 2035 |
+
"grad_norm": 0.0,
|
| 2036 |
+
"kl": 0.0,
|
| 2037 |
+
"learning_rate": 2.0253768788437205e-06,
|
| 2038 |
+
"loss": 0.0007516825199127197,
|
| 2039 |
+
"num_tokens": 3437921.0,
|
| 2040 |
+
"reward": -7.450580596923828e-09,
|
| 2041 |
+
"reward_std": 0.35597237572073936,
|
| 2042 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 7.450580596923828e-09,
|
| 2043 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 2044 |
+
"step": 3650
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"clip_ratio/high_max": 0.0,
|
| 2048 |
+
"clip_ratio/high_mean": 0.0,
|
| 2049 |
+
"clip_ratio/low_mean": 0.0003149054571986198,
|
| 2050 |
+
"clip_ratio/low_min": 0.0003149054571986198,
|
| 2051 |
+
"clip_ratio/region_mean": 0.0003149054571986198,
|
| 2052 |
+
"completion_length": 68.325,
|
| 2053 |
+
"completions/clipped_ratio": 0.0,
|
| 2054 |
+
"completions/max_length": 105.33333333333333,
|
| 2055 |
+
"completions/max_terminated_length": 105.33333333333333,
|
| 2056 |
+
"completions/mean_length": 70.70833333333333,
|
| 2057 |
+
"completions/mean_terminated_length": 70.70833333333333,
|
| 2058 |
+
"completions/min_length": 37.333333333333336,
|
| 2059 |
+
"completions/min_terminated_length": 37.333333333333336,
|
| 2060 |
+
"epoch": 0.006571842673638741,
|
| 2061 |
+
"frac_reward_zero_std": 0.5,
|
| 2062 |
+
"grad_norm": 0.00378933222964406,
|
| 2063 |
+
"kl": 0.0,
|
| 2064 |
+
"learning_rate": 2.0531293710175177e-06,
|
| 2065 |
+
"loss": -0.0022626236081123353,
|
| 2066 |
+
"num_tokens": 3485697.0,
|
| 2067 |
+
"reward": 0.0,
|
| 2068 |
+
"reward_std": 0.4469610129793485,
|
| 2069 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 2.483526865641276e-09,
|
| 2070 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 2071 |
+
"step": 3700
|
| 2072 |
+
},
|
| 2073 |
+
{
|
| 2074 |
+
"clip_ratio/high_max": 0.0001479674782603979,
|
| 2075 |
+
"clip_ratio/high_mean": 0.0001479674782603979,
|
| 2076 |
+
"clip_ratio/low_mean": 0.0003453323198482394,
|
| 2077 |
+
"clip_ratio/low_min": 0.0003453323198482394,
|
| 2078 |
+
"clip_ratio/region_mean": 0.0004932997981086373,
|
| 2079 |
+
"completion_length": 71.745,
|
| 2080 |
+
"completions/clipped_ratio": 0.0,
|
| 2081 |
+
"completions/max_length": 114.0,
|
| 2082 |
+
"completions/max_terminated_length": 114.0,
|
| 2083 |
+
"completions/mean_length": 71.75,
|
| 2084 |
+
"completions/mean_terminated_length": 71.75,
|
| 2085 |
+
"completions/min_length": 48.333333333333336,
|
| 2086 |
+
"completions/min_terminated_length": 48.333333333333336,
|
| 2087 |
+
"epoch": 0.006660651358417642,
|
| 2088 |
+
"frac_reward_zero_std": 0.625,
|
| 2089 |
+
"grad_norm": 0.016572780907154083,
|
| 2090 |
+
"kl": 0.0,
|
| 2091 |
+
"learning_rate": 2.080881863191315e-06,
|
| 2092 |
+
"loss": 0.0012207261472940445,
|
| 2093 |
+
"num_tokens": 3533681.0,
|
| 2094 |
+
"reward": -1.241763432820638e-08,
|
| 2095 |
+
"reward_std": 0.3200628360112508,
|
| 2096 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -9.934107462565104e-09,
|
| 2097 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 2098 |
+
"step": 3750
|
| 2099 |
+
},
|
| 2100 |
+
{
|
| 2101 |
+
"clip_ratio/high_max": 0.0,
|
| 2102 |
+
"clip_ratio/high_mean": 0.0,
|
| 2103 |
+
"clip_ratio/low_mean": 0.000263438718393445,
|
| 2104 |
+
"clip_ratio/low_min": 0.000263438718393445,
|
| 2105 |
+
"clip_ratio/region_mean": 0.000263438718393445,
|
| 2106 |
+
"completion_length": 66.23,
|
| 2107 |
+
"completions/clipped_ratio": 0.0,
|
| 2108 |
+
"completions/max_length": 100.33333333333333,
|
| 2109 |
+
"completions/max_terminated_length": 100.33333333333333,
|
| 2110 |
+
"completions/mean_length": 65.578125,
|
| 2111 |
+
"completions/mean_terminated_length": 65.578125,
|
| 2112 |
+
"completions/min_length": 37.333333333333336,
|
| 2113 |
+
"completions/min_terminated_length": 37.333333333333336,
|
| 2114 |
+
"epoch": 0.006749460043196544,
|
| 2115 |
+
"frac_reward_zero_std": 0.75,
|
| 2116 |
+
"grad_norm": 0.0,
|
| 2117 |
+
"kl": 0.0,
|
| 2118 |
+
"learning_rate": 2.108634355365112e-06,
|
| 2119 |
+
"loss": 0.0009784241765737533,
|
| 2120 |
+
"num_tokens": 3578920.0,
|
| 2121 |
+
"reward": -7.450580596923828e-09,
|
| 2122 |
+
"reward_std": 0.45011167724927265,
|
| 2123 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -4.967053731282552e-09,
|
| 2124 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 2125 |
+
"step": 3800
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"clip_ratio/high_max": 7.352941203862428e-05,
|
| 2129 |
+
"clip_ratio/high_mean": 7.352941203862428e-05,
|
| 2130 |
+
"clip_ratio/low_mean": 5.681818351149559e-05,
|
| 2131 |
+
"clip_ratio/low_min": 5.681818351149559e-05,
|
| 2132 |
+
"clip_ratio/region_mean": 0.00013034759555011988,
|
| 2133 |
+
"completion_length": 66.97,
|
| 2134 |
+
"completions/clipped_ratio": 0.0,
|
| 2135 |
+
"completions/max_length": 98.0,
|
| 2136 |
+
"completions/max_terminated_length": 98.0,
|
| 2137 |
+
"completions/mean_length": 68.484375,
|
| 2138 |
+
"completions/mean_terminated_length": 68.484375,
|
| 2139 |
+
"completions/min_length": 38.333333333333336,
|
| 2140 |
+
"completions/min_terminated_length": 38.333333333333336,
|
| 2141 |
+
"epoch": 0.006838268727975446,
|
| 2142 |
+
"frac_reward_zero_std": 0.7083333333333334,
|
| 2143 |
+
"grad_norm": 0.0,
|
| 2144 |
+
"kl": 0.0,
|
| 2145 |
+
"learning_rate": 2.1363868475389093e-06,
|
| 2146 |
+
"loss": -0.0016198185086250304,
|
| 2147 |
+
"num_tokens": 3626085.0,
|
| 2148 |
+
"reward": 2.60770320892334e-08,
|
| 2149 |
+
"reward_std": 0.28504278759161633,
|
| 2150 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 1.3659397761027018e-08,
|
| 2151 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 2152 |
+
"step": 3850
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"clip_ratio/high_max": 0.00042313845828175547,
|
| 2156 |
+
"clip_ratio/high_mean": 0.00042313845828175547,
|
| 2157 |
+
"clip_ratio/low_mean": 0.0005261846724897623,
|
| 2158 |
+
"clip_ratio/low_min": 0.0005261846724897623,
|
| 2159 |
+
"clip_ratio/region_mean": 0.0009493231307715177,
|
| 2160 |
+
"completion_length": 67.03,
|
| 2161 |
+
"completions/clipped_ratio": 0.0,
|
| 2162 |
+
"completions/max_length": 111.66666666666667,
|
| 2163 |
+
"completions/max_terminated_length": 111.66666666666667,
|
| 2164 |
+
"completions/mean_length": 66.421875,
|
| 2165 |
+
"completions/mean_terminated_length": 66.421875,
|
| 2166 |
+
"completions/min_length": 34.0,
|
| 2167 |
+
"completions/min_terminated_length": 34.0,
|
| 2168 |
+
"epoch": 0.006927077412754348,
|
| 2169 |
+
"frac_reward_zero_std": 0.5833333333333334,
|
| 2170 |
+
"grad_norm": 0.0,
|
| 2171 |
+
"kl": 0.0,
|
| 2172 |
+
"learning_rate": 2.1641393397127065e-06,
|
| 2173 |
+
"loss": 0.0009964856505393983,
|
| 2174 |
+
"num_tokens": 3671542.0,
|
| 2175 |
+
"reward": -7.450580596923828e-09,
|
| 2176 |
+
"reward_std": 0.4345060686270396,
|
| 2177 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -1.241763432820638e-08,
|
| 2178 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 2179 |
+
"step": 3900
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"clip_ratio/high_max": 0.0,
|
| 2183 |
+
"clip_ratio/high_mean": 0.0,
|
| 2184 |
+
"clip_ratio/low_mean": 0.0001675077062100172,
|
| 2185 |
+
"clip_ratio/low_min": 0.0001675077062100172,
|
| 2186 |
+
"clip_ratio/region_mean": 0.0001675077062100172,
|
| 2187 |
+
"completion_length": 64.405,
|
| 2188 |
+
"completions/clipped_ratio": 0.0,
|
| 2189 |
+
"completions/max_length": 124.33333333333333,
|
| 2190 |
+
"completions/max_terminated_length": 124.33333333333333,
|
| 2191 |
+
"completions/mean_length": 64.25520833333333,
|
| 2192 |
+
"completions/mean_terminated_length": 64.25520833333333,
|
| 2193 |
+
"completions/min_length": 26.333333333333332,
|
| 2194 |
+
"completions/min_terminated_length": 26.333333333333332,
|
| 2195 |
+
"epoch": 0.00701588609753325,
|
| 2196 |
+
"frac_reward_zero_std": 0.75,
|
| 2197 |
+
"grad_norm": 0.016584180295467377,
|
| 2198 |
+
"kl": 0.0,
|
| 2199 |
+
"learning_rate": 2.1918918318865037e-06,
|
| 2200 |
+
"loss": 0.00043438874185085295,
|
| 2201 |
+
"num_tokens": 3715719.0,
|
| 2202 |
+
"reward": -2.483526865641276e-09,
|
| 2203 |
+
"reward_std": 0.3461955636739731,
|
| 2204 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": 9.934107462565104e-09,
|
| 2205 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 2206 |
+
"step": 3950
|
| 2207 |
+
},
|
| 2208 |
+
{
|
| 2209 |
+
"clip_ratio/high_max": 8.47457628697157e-05,
|
| 2210 |
+
"clip_ratio/high_mean": 8.47457628697157e-05,
|
| 2211 |
+
"clip_ratio/low_mean": 0.0004254490882158279,
|
| 2212 |
+
"clip_ratio/low_min": 0.0004254490882158279,
|
| 2213 |
+
"clip_ratio/region_mean": 0.0005101948510855437,
|
| 2214 |
+
"completion_length": 64.6,
|
| 2215 |
+
"completions/clipped_ratio": 0.0,
|
| 2216 |
+
"completions/max_length": 120.66666666666667,
|
| 2217 |
+
"completions/max_terminated_length": 120.66666666666667,
|
| 2218 |
+
"completions/mean_length": 64.703125,
|
| 2219 |
+
"completions/mean_terminated_length": 64.703125,
|
| 2220 |
+
"completions/min_length": 22.0,
|
| 2221 |
+
"completions/min_terminated_length": 22.0,
|
| 2222 |
+
"epoch": 0.007104694782312152,
|
| 2223 |
+
"frac_reward_zero_std": 0.7083333333333334,
|
| 2224 |
+
"grad_norm": 0.0,
|
| 2225 |
+
"kl": 0.0,
|
| 2226 |
+
"learning_rate": 2.219644324060301e-06,
|
| 2227 |
+
"loss": -0.0009985111653804779,
|
| 2228 |
+
"num_tokens": 3760262.0,
|
| 2229 |
+
"reward": 2.483526865641276e-09,
|
| 2230 |
+
"reward_std": 0.3999728858470917,
|
| 2231 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -4.967053731282552e-09,
|
| 2232 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 2233 |
+
"step": 4000
|
| 2234 |
+
},
|
| 2235 |
+
{
|
| 2236 |
+
"clip_ratio/high_max": 0.00031662291381508113,
|
| 2237 |
+
"clip_ratio/high_mean": 0.00031662291381508113,
|
| 2238 |
+
"clip_ratio/low_mean": 0.0005780183942988515,
|
| 2239 |
+
"clip_ratio/low_min": 0.0005780183942988515,
|
| 2240 |
+
"clip_ratio/region_mean": 0.0008946413081139326,
|
| 2241 |
+
"completion_length": 69.7,
|
| 2242 |
+
"completions/clipped_ratio": 0.0,
|
| 2243 |
+
"completions/max_length": 119.0,
|
| 2244 |
+
"completions/max_terminated_length": 119.0,
|
| 2245 |
+
"completions/mean_length": 69.3125,
|
| 2246 |
+
"completions/mean_terminated_length": 69.3125,
|
| 2247 |
+
"completions/min_length": 28.0,
|
| 2248 |
+
"completions/min_terminated_length": 28.0,
|
| 2249 |
+
"epoch": 0.007193503467091054,
|
| 2250 |
+
"frac_reward_zero_std": 0.65625,
|
| 2251 |
+
"grad_norm": 0.0009224224486388266,
|
| 2252 |
+
"kl": 0.0,
|
| 2253 |
+
"learning_rate": 2.247396816234098e-06,
|
| 2254 |
+
"loss": 0.000695338100194931,
|
| 2255 |
+
"num_tokens": 3823038.0,
|
| 2256 |
+
"reward": -9.313225746154785e-10,
|
| 2257 |
+
"reward_std": 0.44107016175985336,
|
| 2258 |
+
"rewards/TranscriptCorrectionGrpoReward/mean": -2.7939677238464355e-09,
|
| 2259 |
+
"rewards/TranscriptCorrectionGrpoReward/std": 1.0079052448272705,
|
| 2260 |
+
"step": 4050
|
| 2261 |
+
},
|
| 2262 |
+
{
|
| 2263 |
+
"eval_cer_subset": 0.01319067858713176,
|
| 2264 |
+
"eval_cer_subset_edit_distance": 810,
|
| 2265 |
+
"eval_cer_subset_groups": 250,
|
| 2266 |
+
"eval_cer_subset_items": 250,
|
| 2267 |
+
"eval_cer_subset_ref_chars": 61407,
|
| 2268 |
+
"eval_loss": 0.0427304402914159,
|
| 2269 |
+
"step": 4056
|
| 2270 |
+
}
|
| 2271 |
+
],
|
| 2272 |
+
"logging_steps": 50,
|
| 2273 |
+
"max_steps": 2252032,
|
| 2274 |
+
"num_input_tokens_seen": 3823038,
|
| 2275 |
+
"num_train_epochs": 4,
|
| 2276 |
+
"save_steps": 507,
|
| 2277 |
+
"stateful_callbacks": {
|
| 2278 |
+
"TrainerControl": {
|
| 2279 |
+
"args": {
|
| 2280 |
+
"should_epoch_stop": false,
|
| 2281 |
+
"should_evaluate": false,
|
| 2282 |
+
"should_log": false,
|
| 2283 |
+
"should_save": true,
|
| 2284 |
+
"should_training_stop": false
|
| 2285 |
+
},
|
| 2286 |
+
"attributes": {}
|
| 2287 |
+
}
|
| 2288 |
+
},
|
| 2289 |
+
"total_flos": 0.0,
|
| 2290 |
+
"train_batch_size": 2,
|
| 2291 |
+
"trial_name": null,
|
| 2292 |
+
"trial_params": null
|
| 2293 |
+
}
|
meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:efd03f5456a4c15d2256ba2960a071c6dc2cb8a2ff8cf43b4f1331b6c112c442
|
| 3 |
+
size 6737
|
meta-llama__llama-3.2-1b/grpo/run_manifest.json
CHANGED
|
@@ -9,7 +9,7 @@
|
|
| 9 |
"best_metric_value": 0.01195303467031446,
|
| 10 |
"best_step": 3042,
|
| 11 |
"best_checkpoint_step": 3042,
|
| 12 |
-
"latest_checkpoint_step":
|
| 13 |
"checkpoints": [
|
| 14 |
{
|
| 15 |
"step": 507,
|
|
@@ -52,10 +52,16 @@
|
|
| 52 |
"epoch": 0.006303640445606456,
|
| 53 |
"repo_path": "meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-3549",
|
| 54 |
"timestamp": "2026-04-09T01:39:39.758317+00:00"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
}
|
| 56 |
],
|
| 57 |
"epoch_artifacts": [],
|
| 58 |
"resume_history": [],
|
| 59 |
"created_at": "2026-04-09T00:54:21.973160+00:00",
|
| 60 |
-
"updated_at": "2026-04-09T01:
|
| 61 |
}
|
|
|
|
| 9 |
"best_metric_value": 0.01195303467031446,
|
| 10 |
"best_step": 3042,
|
| 11 |
"best_checkpoint_step": 3042,
|
| 12 |
+
"latest_checkpoint_step": 4056,
|
| 13 |
"checkpoints": [
|
| 14 |
{
|
| 15 |
"step": 507,
|
|
|
|
| 52 |
"epoch": 0.006303640445606456,
|
| 53 |
"repo_path": "meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-3549",
|
| 54 |
"timestamp": "2026-04-09T01:39:39.758317+00:00"
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"step": 4056,
|
| 58 |
+
"epoch": 0.007204160509264522,
|
| 59 |
+
"repo_path": "meta-llama__llama-3.2-1b/grpo/checkpoints/checkpoint-4056",
|
| 60 |
+
"timestamp": "2026-04-09T01:44:53.215161+00:00"
|
| 61 |
}
|
| 62 |
],
|
| 63 |
"epoch_artifacts": [],
|
| 64 |
"resume_history": [],
|
| 65 |
"created_at": "2026-04-09T00:54:21.973160+00:00",
|
| 66 |
+
"updated_at": "2026-04-09T01:44:53.215176+00:00"
|
| 67 |
}
|