Text Generation
Transformers
Safetensors
English
qwen3
text-generation-inference
unsloth
conversational
Instructions to use CompassioninMachineLearning/qwen-compassion-web-s3407-part1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use CompassioninMachineLearning/qwen-compassion-web-s3407-part1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="CompassioninMachineLearning/qwen-compassion-web-s3407-part1") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("CompassioninMachineLearning/qwen-compassion-web-s3407-part1") model = AutoModelForCausalLM.from_pretrained("CompassioninMachineLearning/qwen-compassion-web-s3407-part1", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use CompassioninMachineLearning/qwen-compassion-web-s3407-part1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "CompassioninMachineLearning/qwen-compassion-web-s3407-part1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "CompassioninMachineLearning/qwen-compassion-web-s3407-part1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/CompassioninMachineLearning/qwen-compassion-web-s3407-part1
- SGLang
How to use CompassioninMachineLearning/qwen-compassion-web-s3407-part1 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "CompassioninMachineLearning/qwen-compassion-web-s3407-part1" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "CompassioninMachineLearning/qwen-compassion-web-s3407-part1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "CompassioninMachineLearning/qwen-compassion-web-s3407-part1" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "CompassioninMachineLearning/qwen-compassion-web-s3407-part1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Unsloth Desktop
- Docker Model Runner
How to use CompassioninMachineLearning/qwen-compassion-web-s3407-part1 with Docker Model Runner:
docker model run hf.co/CompassioninMachineLearning/qwen-compassion-web-s3407-part1
add trainer_state.json
Browse files- trainer_state.json +993 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,993 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 600,
|
| 3 |
+
"best_metric": 0.2658693790435791,
|
| 4 |
+
"best_model_checkpoint": "/workspace/anchor/qwen-compassion-web-s3407/train/checkpoint-600",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 614,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.008143322475570033,
|
| 14 |
+
"grad_norm": 8.803375244140625,
|
| 15 |
+
"learning_rate": 2.5806451612903226e-05,
|
| 16 |
+
"loss": 1.2056,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.016286644951140065,
|
| 21 |
+
"grad_norm": 3.7129921913146973,
|
| 22 |
+
"learning_rate": 5.8064516129032266e-05,
|
| 23 |
+
"loss": 0.9384,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.024429967426710098,
|
| 28 |
+
"grad_norm": 2.6070480346679688,
|
| 29 |
+
"learning_rate": 9.032258064516129e-05,
|
| 30 |
+
"loss": 0.8639,
|
| 31 |
+
"step": 15
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03257328990228013,
|
| 35 |
+
"grad_norm": 2.043976068496704,
|
| 36 |
+
"learning_rate": 0.00012258064516129034,
|
| 37 |
+
"loss": 0.82,
|
| 38 |
+
"step": 20
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04071661237785016,
|
| 42 |
+
"grad_norm": 3.751563787460327,
|
| 43 |
+
"learning_rate": 0.00015483870967741937,
|
| 44 |
+
"loss": 0.7608,
|
| 45 |
+
"step": 25
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.048859934853420196,
|
| 49 |
+
"grad_norm": 2.8326261043548584,
|
| 50 |
+
"learning_rate": 0.0001870967741935484,
|
| 51 |
+
"loss": 0.6016,
|
| 52 |
+
"step": 30
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.057003257328990226,
|
| 56 |
+
"grad_norm": 2.2605817317962646,
|
| 57 |
+
"learning_rate": 0.00019998693330688282,
|
| 58 |
+
"loss": 0.6963,
|
| 59 |
+
"step": 35
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.06514657980456026,
|
| 63 |
+
"grad_norm": 2.8328440189361572,
|
| 64 |
+
"learning_rate": 0.00019990709365929677,
|
| 65 |
+
"loss": 0.5558,
|
| 66 |
+
"step": 40
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0732899022801303,
|
| 70 |
+
"grad_norm": 2.2280285358428955,
|
| 71 |
+
"learning_rate": 0.00019975473152335726,
|
| 72 |
+
"loss": 0.5615,
|
| 73 |
+
"step": 45
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.08143322475570032,
|
| 77 |
+
"grad_norm": 1.7250114679336548,
|
| 78 |
+
"learning_rate": 0.0001995299574986306,
|
| 79 |
+
"loss": 0.5391,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.08143322475570032,
|
| 84 |
+
"eval_loss": 0.374595582485199,
|
| 85 |
+
"eval_runtime": 163.9265,
|
| 86 |
+
"eval_samples_per_second": 1.537,
|
| 87 |
+
"eval_steps_per_second": 0.769,
|
| 88 |
+
"step": 50
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.08957654723127036,
|
| 92 |
+
"grad_norm": 1.6075690984725952,
|
| 93 |
+
"learning_rate": 0.00019923293474842174,
|
| 94 |
+
"loss": 0.5888,
|
| 95 |
+
"step": 55
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.09771986970684039,
|
| 99 |
+
"grad_norm": 1.2532232999801636,
|
| 100 |
+
"learning_rate": 0.00019886387888133413,
|
| 101 |
+
"loss": 0.4921,
|
| 102 |
+
"step": 60
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.10586319218241043,
|
| 106 |
+
"grad_norm": 1.2344268560409546,
|
| 107 |
+
"learning_rate": 0.00019842305779475968,
|
| 108 |
+
"loss": 0.3895,
|
| 109 |
+
"step": 65
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.11400651465798045,
|
| 113 |
+
"grad_norm": 1.2183916568756104,
|
| 114 |
+
"learning_rate": 0.0001979107914804122,
|
| 115 |
+
"loss": 0.5535,
|
| 116 |
+
"step": 70
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.12214983713355049,
|
| 120 |
+
"grad_norm": 0.927330493927002,
|
| 121 |
+
"learning_rate": 0.00019732745179204552,
|
| 122 |
+
"loss": 0.4052,
|
| 123 |
+
"step": 75
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.13029315960912052,
|
| 127 |
+
"grad_norm": 1.6917295455932617,
|
| 128 |
+
"learning_rate": 0.00019667346217552527,
|
| 129 |
+
"loss": 0.4075,
|
| 130 |
+
"step": 80
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.13843648208469056,
|
| 134 |
+
"grad_norm": 1.3769302368164062,
|
| 135 |
+
"learning_rate": 0.00019594929736144976,
|
| 136 |
+
"loss": 0.5872,
|
| 137 |
+
"step": 85
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.1465798045602606,
|
| 141 |
+
"grad_norm": 1.2509673833847046,
|
| 142 |
+
"learning_rate": 0.00019515548302054335,
|
| 143 |
+
"loss": 0.4777,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.15472312703583063,
|
| 148 |
+
"grad_norm": 1.104792594909668,
|
| 149 |
+
"learning_rate": 0.0001942925953820725,
|
| 150 |
+
"loss": 0.4456,
|
| 151 |
+
"step": 95
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.16286644951140064,
|
| 155 |
+
"grad_norm": 1.119893193244934,
|
| 156 |
+
"learning_rate": 0.00019336126081556134,
|
| 157 |
+
"loss": 0.5064,
|
| 158 |
+
"step": 100
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.16286644951140064,
|
| 162 |
+
"eval_loss": 0.3101080060005188,
|
| 163 |
+
"eval_runtime": 83.9108,
|
| 164 |
+
"eval_samples_per_second": 3.003,
|
| 165 |
+
"eval_steps_per_second": 1.502,
|
| 166 |
+
"step": 100
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.17100977198697068,
|
| 170 |
+
"grad_norm": 1.2182846069335938,
|
| 171 |
+
"learning_rate": 0.00019236215537611046,
|
| 172 |
+
"loss": 0.4342,
|
| 173 |
+
"step": 105
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.1791530944625407,
|
| 177 |
+
"grad_norm": 1.2241324186325073,
|
| 178 |
+
"learning_rate": 0.00019129600431364897,
|
| 179 |
+
"loss": 0.4511,
|
| 180 |
+
"step": 110
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.18729641693811075,
|
| 184 |
+
"grad_norm": 1.2518731355667114,
|
| 185 |
+
"learning_rate": 0.00019016358154647618,
|
| 186 |
+
"loss": 0.4454,
|
| 187 |
+
"step": 115
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.19543973941368079,
|
| 191 |
+
"grad_norm": 1.3357152938842773,
|
| 192 |
+
"learning_rate": 0.00018896570909947475,
|
| 193 |
+
"loss": 0.4239,
|
| 194 |
+
"step": 120
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.20358306188925082,
|
| 198 |
+
"grad_norm": 1.1552698612213135,
|
| 199 |
+
"learning_rate": 0.00018770325650740345,
|
| 200 |
+
"loss": 0.418,
|
| 201 |
+
"step": 125
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.21172638436482086,
|
| 205 |
+
"grad_norm": 1.177544355392456,
|
| 206 |
+
"learning_rate": 0.00018637714018370253,
|
| 207 |
+
"loss": 0.4731,
|
| 208 |
+
"step": 130
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.21986970684039087,
|
| 212 |
+
"grad_norm": 0.8246702551841736,
|
| 213 |
+
"learning_rate": 0.00018498832275526988,
|
| 214 |
+
"loss": 0.4305,
|
| 215 |
+
"step": 135
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.2280130293159609,
|
| 219 |
+
"grad_norm": 0.6699772477149963,
|
| 220 |
+
"learning_rate": 0.00018353781236369064,
|
| 221 |
+
"loss": 0.4319,
|
| 222 |
+
"step": 140
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.23615635179153094,
|
| 226 |
+
"grad_norm": 1.1820833683013916,
|
| 227 |
+
"learning_rate": 0.00018202666193342833,
|
| 228 |
+
"loss": 0.4795,
|
| 229 |
+
"step": 145
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.24429967426710097,
|
| 233 |
+
"grad_norm": 1.3192893266677856,
|
| 234 |
+
"learning_rate": 0.00018045596840750723,
|
| 235 |
+
"loss": 0.4758,
|
| 236 |
+
"step": 150
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.24429967426710097,
|
| 240 |
+
"eval_loss": 0.2977350652217865,
|
| 241 |
+
"eval_runtime": 83.4415,
|
| 242 |
+
"eval_samples_per_second": 3.02,
|
| 243 |
+
"eval_steps_per_second": 1.51,
|
| 244 |
+
"step": 150
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.252442996742671,
|
| 248 |
+
"grad_norm": 0.9550138711929321,
|
| 249 |
+
"learning_rate": 0.0001788268719512427,
|
| 250 |
+
"loss": 0.4393,
|
| 251 |
+
"step": 155
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.26058631921824105,
|
| 255 |
+
"grad_norm": 1.382379174232483,
|
| 256 |
+
"learning_rate": 0.00017714055512459565,
|
| 257 |
+
"loss": 0.4298,
|
| 258 |
+
"step": 160
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.2687296416938111,
|
| 262 |
+
"grad_norm": 0.9943409562110901,
|
| 263 |
+
"learning_rate": 0.0001753982420237533,
|
| 264 |
+
"loss": 0.5009,
|
| 265 |
+
"step": 165
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.2768729641693811,
|
| 269 |
+
"grad_norm": 1.2051012516021729,
|
| 270 |
+
"learning_rate": 0.00017360119739255852,
|
| 271 |
+
"loss": 0.3328,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.28501628664495116,
|
| 276 |
+
"grad_norm": 0.7656549215316772,
|
| 277 |
+
"learning_rate": 0.00017175072570443312,
|
| 278 |
+
"loss": 0.4809,
|
| 279 |
+
"step": 175
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.2931596091205212,
|
| 283 |
+
"grad_norm": 0.8127760291099548,
|
| 284 |
+
"learning_rate": 0.00016984817021546177,
|
| 285 |
+
"loss": 0.37,
|
| 286 |
+
"step": 180
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.30130293159609123,
|
| 290 |
+
"grad_norm": 1.5806375741958618,
|
| 291 |
+
"learning_rate": 0.000167894911989323,
|
| 292 |
+
"loss": 0.4262,
|
| 293 |
+
"step": 185
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 0.30944625407166126,
|
| 297 |
+
"grad_norm": 1.1315789222717285,
|
| 298 |
+
"learning_rate": 0.00016589236889477646,
|
| 299 |
+
"loss": 0.3342,
|
| 300 |
+
"step": 190
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 0.31758957654723124,
|
| 304 |
+
"grad_norm": 1.2172818183898926,
|
| 305 |
+
"learning_rate": 0.00016384199457643262,
|
| 306 |
+
"loss": 0.4647,
|
| 307 |
+
"step": 195
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 0.3257328990228013,
|
| 311 |
+
"grad_norm": 1.0508732795715332,
|
| 312 |
+
"learning_rate": 0.00016174527739955342,
|
| 313 |
+
"loss": 0.5813,
|
| 314 |
+
"step": 200
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.3257328990228013,
|
| 318 |
+
"eval_loss": 0.29167938232421875,
|
| 319 |
+
"eval_runtime": 83.6844,
|
| 320 |
+
"eval_samples_per_second": 3.011,
|
| 321 |
+
"eval_steps_per_second": 1.506,
|
| 322 |
+
"step": 200
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 0.3338762214983713,
|
| 326 |
+
"grad_norm": 0.6504319310188293,
|
| 327 |
+
"learning_rate": 0.00015960373936964892,
|
| 328 |
+
"loss": 0.385,
|
| 329 |
+
"step": 205
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.34201954397394135,
|
| 333 |
+
"grad_norm": 1.1721259355545044,
|
| 334 |
+
"learning_rate": 0.0001574189350276545,
|
| 335 |
+
"loss": 0.4746,
|
| 336 |
+
"step": 210
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.3501628664495114,
|
| 340 |
+
"grad_norm": 1.2938706874847412,
|
| 341 |
+
"learning_rate": 0.00015519245032149083,
|
| 342 |
+
"loss": 0.5205,
|
| 343 |
+
"step": 215
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 0.3583061889250814,
|
| 347 |
+
"grad_norm": 0.9688608646392822,
|
| 348 |
+
"learning_rate": 0.0001529259014548253,
|
| 349 |
+
"loss": 0.4305,
|
| 350 |
+
"step": 220
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.36644951140065146,
|
| 354 |
+
"grad_norm": 1.1749860048294067,
|
| 355 |
+
"learning_rate": 0.00015062093371387097,
|
| 356 |
+
"loss": 0.4359,
|
| 357 |
+
"step": 225
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.3745928338762215,
|
| 361 |
+
"grad_norm": 1.3086587190628052,
|
| 362 |
+
"learning_rate": 0.00014827922027307451,
|
| 363 |
+
"loss": 0.4028,
|
| 364 |
+
"step": 230
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 0.38273615635179153,
|
| 368 |
+
"grad_norm": 1.0187360048294067,
|
| 369 |
+
"learning_rate": 0.00014590246098055996,
|
| 370 |
+
"loss": 0.4615,
|
| 371 |
+
"step": 235
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 0.39087947882736157,
|
| 375 |
+
"grad_norm": 0.8567041754722595,
|
| 376 |
+
"learning_rate": 0.00014349238112421024,
|
| 377 |
+
"loss": 0.3113,
|
| 378 |
+
"step": 240
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 0.3990228013029316,
|
| 382 |
+
"grad_norm": 0.9721431136131287,
|
| 383 |
+
"learning_rate": 0.000141050730179282,
|
| 384 |
+
"loss": 0.4088,
|
| 385 |
+
"step": 245
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.40716612377850164,
|
| 389 |
+
"grad_norm": 1.1704206466674805,
|
| 390 |
+
"learning_rate": 0.0001385792805384625,
|
| 391 |
+
"loss": 0.5045,
|
| 392 |
+
"step": 250
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.40716612377850164,
|
| 396 |
+
"eval_loss": 0.2854733467102051,
|
| 397 |
+
"eval_runtime": 83.6144,
|
| 398 |
+
"eval_samples_per_second": 3.014,
|
| 399 |
+
"eval_steps_per_second": 1.507,
|
| 400 |
+
"step": 250
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.4153094462540717,
|
| 404 |
+
"grad_norm": 1.115260124206543,
|
| 405 |
+
"learning_rate": 0.00013607982622529133,
|
| 406 |
+
"loss": 0.4098,
|
| 407 |
+
"step": 255
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.4234527687296417,
|
| 411 |
+
"grad_norm": 1.2716691493988037,
|
| 412 |
+
"learning_rate": 0.00013355418159187985,
|
| 413 |
+
"loss": 0.4753,
|
| 414 |
+
"step": 260
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.43159609120521175,
|
| 418 |
+
"grad_norm": 0.6622514724731445,
|
| 419 |
+
"learning_rate": 0.00013100418000187419,
|
| 420 |
+
"loss": 0.4589,
|
| 421 |
+
"step": 265
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.43973941368078173,
|
| 425 |
+
"grad_norm": 1.056097149848938,
|
| 426 |
+
"learning_rate": 0.0001284316724996181,
|
| 427 |
+
"loss": 0.4588,
|
| 428 |
+
"step": 270
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.44788273615635177,
|
| 432 |
+
"grad_norm": 1.1712679862976074,
|
| 433 |
+
"learning_rate": 0.00012583852646648095,
|
| 434 |
+
"loss": 0.5739,
|
| 435 |
+
"step": 275
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 0.4560260586319218,
|
| 439 |
+
"grad_norm": 1.0975291728973389,
|
| 440 |
+
"learning_rate": 0.00012322662426532708,
|
| 441 |
+
"loss": 0.4602,
|
| 442 |
+
"step": 280
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.46416938110749184,
|
| 446 |
+
"grad_norm": 0.7166972756385803,
|
| 447 |
+
"learning_rate": 0.00012059786187410984,
|
| 448 |
+
"loss": 0.3027,
|
| 449 |
+
"step": 285
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.4723127035830619,
|
| 453 |
+
"grad_norm": 0.63871830701828,
|
| 454 |
+
"learning_rate": 0.00011795414750958265,
|
| 455 |
+
"loss": 0.2792,
|
| 456 |
+
"step": 290
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.4804560260586319,
|
| 460 |
+
"grad_norm": 1.0939736366271973,
|
| 461 |
+
"learning_rate": 0.00011529740024212565,
|
| 462 |
+
"loss": 0.4039,
|
| 463 |
+
"step": 295
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 0.48859934853420195,
|
| 467 |
+
"grad_norm": 1.2159818410873413,
|
| 468 |
+
"learning_rate": 0.00011262954860269399,
|
| 469 |
+
"loss": 0.5488,
|
| 470 |
+
"step": 300
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.48859934853420195,
|
| 474 |
+
"eval_loss": 0.2815026044845581,
|
| 475 |
+
"eval_runtime": 84.5088,
|
| 476 |
+
"eval_samples_per_second": 2.982,
|
| 477 |
+
"eval_steps_per_second": 1.491,
|
| 478 |
+
"step": 300
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.496742671009772,
|
| 482 |
+
"grad_norm": 1.18336021900177,
|
| 483 |
+
"learning_rate": 0.0001099525291828986,
|
| 484 |
+
"loss": 0.3673,
|
| 485 |
+
"step": 305
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.504885993485342,
|
| 489 |
+
"grad_norm": 1.2917401790618896,
|
| 490 |
+
"learning_rate": 0.00010726828522923562,
|
| 491 |
+
"loss": 0.402,
|
| 492 |
+
"step": 310
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.5130293159609121,
|
| 496 |
+
"grad_norm": 1.1676180362701416,
|
| 497 |
+
"learning_rate": 0.00010457876523248518,
|
| 498 |
+
"loss": 0.4644,
|
| 499 |
+
"step": 315
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.5211726384364821,
|
| 503 |
+
"grad_norm": 0.6200851202011108,
|
| 504 |
+
"learning_rate": 0.00010188592151330343,
|
| 505 |
+
"loss": 0.3415,
|
| 506 |
+
"step": 320
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.5293159609120521,
|
| 510 |
+
"grad_norm": 0.6668900847434998,
|
| 511 |
+
"learning_rate": 9.919170880503415e-05,
|
| 512 |
+
"loss": 0.4088,
|
| 513 |
+
"step": 325
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.5374592833876222,
|
| 517 |
+
"grad_norm": 1.1132478713989258,
|
| 518 |
+
"learning_rate": 9.649808283476941e-05,
|
| 519 |
+
"loss": 0.2441,
|
| 520 |
+
"step": 330
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.5456026058631922,
|
| 524 |
+
"grad_norm": 0.7450593113899231,
|
| 525 |
+
"learning_rate": 9.38069989036886e-05,
|
| 526 |
+
"loss": 0.4173,
|
| 527 |
+
"step": 335
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.5537459283387622,
|
| 531 |
+
"grad_norm": 0.7038878798484802,
|
| 532 |
+
"learning_rate": 9.112041046770653e-05,
|
| 533 |
+
"loss": 0.2244,
|
| 534 |
+
"step": 340
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.5618892508143323,
|
| 538 |
+
"grad_norm": 1.1429181098937988,
|
| 539 |
+
"learning_rate": 8.844026771946147e-05,
|
| 540 |
+
"loss": 0.4247,
|
| 541 |
+
"step": 345
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.5700325732899023,
|
| 545 |
+
"grad_norm": 1.2036070823669434,
|
| 546 |
+
"learning_rate": 8.57685161726715e-05,
|
| 547 |
+
"loss": 0.4933,
|
| 548 |
+
"step": 350
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.5700325732899023,
|
| 552 |
+
"eval_loss": 0.2762308418750763,
|
| 553 |
+
"eval_runtime": 83.4746,
|
| 554 |
+
"eval_samples_per_second": 3.019,
|
| 555 |
+
"eval_steps_per_second": 1.509,
|
| 556 |
+
"step": 350
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.5781758957654723,
|
| 560 |
+
"grad_norm": 0.9277458786964417,
|
| 561 |
+
"learning_rate": 8.310709524988743e-05,
|
| 562 |
+
"loss": 0.4341,
|
| 563 |
+
"step": 355
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.5863192182410424,
|
| 567 |
+
"grad_norm": 1.0389032363891602,
|
| 568 |
+
"learning_rate": 8.045793687466757e-05,
|
| 569 |
+
"loss": 0.327,
|
| 570 |
+
"step": 360
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.5944625407166124,
|
| 574 |
+
"grad_norm": 1.003150463104248,
|
| 575 |
+
"learning_rate": 7.782296406919557e-05,
|
| 576 |
+
"loss": 0.4683,
|
| 577 |
+
"step": 365
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.6026058631921825,
|
| 581 |
+
"grad_norm": 1.0817468166351318,
|
| 582 |
+
"learning_rate": 7.520408955836007e-05,
|
| 583 |
+
"loss": 0.2979,
|
| 584 |
+
"step": 370
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6107491856677525,
|
| 588 |
+
"grad_norm": 0.6353104114532471,
|
| 589 |
+
"learning_rate": 7.260321438130913e-05,
|
| 590 |
+
"loss": 0.3314,
|
| 591 |
+
"step": 375
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.6188925081433225,
|
| 595 |
+
"grad_norm": 0.8815907835960388,
|
| 596 |
+
"learning_rate": 7.002222651148714e-05,
|
| 597 |
+
"loss": 0.2524,
|
| 598 |
+
"step": 380
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.6270358306188925,
|
| 602 |
+
"grad_norm": 1.3402814865112305,
|
| 603 |
+
"learning_rate": 6.746299948615631e-05,
|
| 604 |
+
"loss": 0.5739,
|
| 605 |
+
"step": 385
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.6351791530944625,
|
| 609 |
+
"grad_norm": 0.8972781896591187,
|
| 610 |
+
"learning_rate": 6.492739104639727e-05,
|
| 611 |
+
"loss": 0.3207,
|
| 612 |
+
"step": 390
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.6433224755700325,
|
| 616 |
+
"grad_norm": 0.8222898840904236,
|
| 617 |
+
"learning_rate": 6.24172417885762e-05,
|
| 618 |
+
"loss": 0.4004,
|
| 619 |
+
"step": 395
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.6514657980456026,
|
| 623 |
+
"grad_norm": 1.0864665508270264,
|
| 624 |
+
"learning_rate": 5.9934373828257106e-05,
|
| 625 |
+
"loss": 0.335,
|
| 626 |
+
"step": 400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.6514657980456026,
|
| 630 |
+
"eval_loss": 0.27289000153541565,
|
| 631 |
+
"eval_runtime": 83.8616,
|
| 632 |
+
"eval_samples_per_second": 3.005,
|
| 633 |
+
"eval_steps_per_second": 1.502,
|
| 634 |
+
"step": 400
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.6596091205211726,
|
| 638 |
+
"grad_norm": 0.620107889175415,
|
| 639 |
+
"learning_rate": 5.7480589477529545e-05,
|
| 640 |
+
"loss": 0.3715,
|
| 641 |
+
"step": 405
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.6677524429967426,
|
| 645 |
+
"grad_norm": 0.9920016527175903,
|
| 646 |
+
"learning_rate": 5.505766993671162e-05,
|
| 647 |
+
"loss": 0.4575,
|
| 648 |
+
"step": 410
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.6758957654723127,
|
| 652 |
+
"grad_norm": 1.1352708339691162,
|
| 653 |
+
"learning_rate": 5.266737400137765e-05,
|
| 654 |
+
"loss": 0.2978,
|
| 655 |
+
"step": 415
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.6840390879478827,
|
| 659 |
+
"grad_norm": 1.303180456161499,
|
| 660 |
+
"learning_rate": 5.031143678565005e-05,
|
| 661 |
+
"loss": 0.3386,
|
| 662 |
+
"step": 420
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.6921824104234527,
|
| 666 |
+
"grad_norm": 0.6123512387275696,
|
| 667 |
+
"learning_rate": 4.799156846268095e-05,
|
| 668 |
+
"loss": 0.3016,
|
| 669 |
+
"step": 425
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.7003257328990228,
|
| 673 |
+
"grad_norm": 1.2777764797210693,
|
| 674 |
+
"learning_rate": 4.57094530232389e-05,
|
| 675 |
+
"loss": 0.4377,
|
| 676 |
+
"step": 430
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.7084690553745928,
|
| 680 |
+
"grad_norm": 0.9900017380714417,
|
| 681 |
+
"learning_rate": 4.346674705330117e-05,
|
| 682 |
+
"loss": 0.4437,
|
| 683 |
+
"step": 435
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.7166123778501629,
|
| 687 |
+
"grad_norm": 1.029066801071167,
|
| 688 |
+
"learning_rate": 4.1265078531538916e-05,
|
| 689 |
+
"loss": 0.3486,
|
| 690 |
+
"step": 440
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.7247557003257329,
|
| 694 |
+
"grad_norm": 1.181535005569458,
|
| 695 |
+
"learning_rate": 3.9106045647569e-05,
|
| 696 |
+
"loss": 0.4141,
|
| 697 |
+
"step": 445
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.7328990228013029,
|
| 701 |
+
"grad_norm": 1.3165513277053833,
|
| 702 |
+
"learning_rate": 3.69912156418289e-05,
|
| 703 |
+
"loss": 0.5187,
|
| 704 |
+
"step": 450
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.7328990228013029,
|
| 708 |
+
"eval_loss": 0.2702295184135437,
|
| 709 |
+
"eval_runtime": 83.3618,
|
| 710 |
+
"eval_samples_per_second": 3.023,
|
| 711 |
+
"eval_steps_per_second": 1.511,
|
| 712 |
+
"step": 450
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.741042345276873,
|
| 716 |
+
"grad_norm": 1.1030694246292114,
|
| 717 |
+
"learning_rate": 3.492212366791831e-05,
|
| 718 |
+
"loss": 0.3816,
|
| 719 |
+
"step": 455
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 0.749185667752443,
|
| 723 |
+
"grad_norm": 1.0479238033294678,
|
| 724 |
+
"learning_rate": 3.290027167823204e-05,
|
| 725 |
+
"loss": 0.3815,
|
| 726 |
+
"step": 460
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 0.757328990228013,
|
| 730 |
+
"grad_norm": 1.0574332475662231,
|
| 731 |
+
"learning_rate": 3.092712733369387e-05,
|
| 732 |
+
"loss": 0.3443,
|
| 733 |
+
"step": 465
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 0.7654723127035831,
|
| 737 |
+
"grad_norm": 0.7265905737876892,
|
| 738 |
+
"learning_rate": 2.9004122938382617e-05,
|
| 739 |
+
"loss": 0.3229,
|
| 740 |
+
"step": 470
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.7736156351791531,
|
| 744 |
+
"grad_norm": 0.9613872766494751,
|
| 745 |
+
"learning_rate": 2.7132654399823444e-05,
|
| 746 |
+
"loss": 0.4289,
|
| 747 |
+
"step": 475
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 0.7817589576547231,
|
| 751 |
+
"grad_norm": 0.8705881834030151,
|
| 752 |
+
"learning_rate": 2.531408021569982e-05,
|
| 753 |
+
"loss": 0.391,
|
| 754 |
+
"step": 480
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.7899022801302932,
|
| 758 |
+
"grad_norm": 0.9562520384788513,
|
| 759 |
+
"learning_rate": 2.3549720487720738e-05,
|
| 760 |
+
"loss": 0.4853,
|
| 761 |
+
"step": 485
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.7980456026058632,
|
| 765 |
+
"grad_norm": 1.064986228942871,
|
| 766 |
+
"learning_rate": 2.184085596336011e-05,
|
| 767 |
+
"loss": 0.4105,
|
| 768 |
+
"step": 490
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.8061889250814332,
|
| 772 |
+
"grad_norm": 1.1484014987945557,
|
| 773 |
+
"learning_rate": 2.0188727106162874e-05,
|
| 774 |
+
"loss": 0.4579,
|
| 775 |
+
"step": 495
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.8143322475570033,
|
| 779 |
+
"grad_norm": 0.9046717286109924,
|
| 780 |
+
"learning_rate": 1.8594533195293427e-05,
|
| 781 |
+
"loss": 0.4128,
|
| 782 |
+
"step": 500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.8143322475570033,
|
| 786 |
+
"eval_loss": 0.26746508479118347,
|
| 787 |
+
"eval_runtime": 83.6362,
|
| 788 |
+
"eval_samples_per_second": 3.013,
|
| 789 |
+
"eval_steps_per_second": 1.507,
|
| 790 |
+
"step": 500
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 0.8224755700325733,
|
| 794 |
+
"grad_norm": 1.0069684982299805,
|
| 795 |
+
"learning_rate": 1.7059431454979824e-05,
|
| 796 |
+
"loss": 0.4211,
|
| 797 |
+
"step": 505
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 0.8306188925081434,
|
| 801 |
+
"grad_norm": 1.418688416481018,
|
| 802 |
+
"learning_rate": 1.5584536214485457e-05,
|
| 803 |
+
"loss": 0.3868,
|
| 804 |
+
"step": 510
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 0.8387622149837134,
|
| 808 |
+
"grad_norm": 1.0533849000930786,
|
| 809 |
+
"learning_rate": 1.4170918099218166e-05,
|
| 810 |
+
"loss": 0.3819,
|
| 811 |
+
"step": 515
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 0.8469055374592834,
|
| 815 |
+
"grad_norm": 0.9637581706047058,
|
| 816 |
+
"learning_rate": 1.2819603253564205e-05,
|
| 817 |
+
"loss": 0.3536,
|
| 818 |
+
"step": 520
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.8550488599348535,
|
| 822 |
+
"grad_norm": 1.0241345167160034,
|
| 823 |
+
"learning_rate": 1.153157259601062e-05,
|
| 824 |
+
"loss": 0.3699,
|
| 825 |
+
"step": 525
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 0.8631921824104235,
|
| 829 |
+
"grad_norm": 0.985444962978363,
|
| 830 |
+
"learning_rate": 1.030776110709718e-05,
|
| 831 |
+
"loss": 0.4363,
|
| 832 |
+
"step": 530
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 0.8713355048859935,
|
| 836 |
+
"grad_norm": 1.0319409370422363,
|
| 837 |
+
"learning_rate": 9.149057150714801e-06,
|
| 838 |
+
"loss": 0.4192,
|
| 839 |
+
"step": 535
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 0.8794788273615635,
|
| 843 |
+
"grad_norm": 1.0504857301712036,
|
| 844 |
+
"learning_rate": 8.056301829242784e-06,
|
| 845 |
+
"loss": 0.3702,
|
| 846 |
+
"step": 540
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 0.8876221498371335,
|
| 850 |
+
"grad_norm": 0.662237286567688,
|
| 851 |
+
"learning_rate": 7.030288372993066e-06,
|
| 852 |
+
"loss": 0.2473,
|
| 853 |
+
"step": 545
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 0.8957654723127035,
|
| 857 |
+
"grad_norm": 1.0782599449157715,
|
| 858 |
+
"learning_rate": 6.0717615644051206e-06,
|
| 859 |
+
"loss": 0.4462,
|
| 860 |
+
"step": 550
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.8957654723127035,
|
| 864 |
+
"eval_loss": 0.2662240266799927,
|
| 865 |
+
"eval_runtime": 83.3987,
|
| 866 |
+
"eval_samples_per_second": 3.022,
|
| 867 |
+
"eval_steps_per_second": 1.511,
|
| 868 |
+
"step": 550
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 0.9039087947882736,
|
| 872 |
+
"grad_norm": 0.8521231412887573,
|
| 873 |
+
"learning_rate": 5.181417197408734e-06,
|
| 874 |
+
"loss": 0.4453,
|
| 875 |
+
"step": 555
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 0.9120521172638436,
|
| 879 |
+
"grad_norm": 0.8748440742492676,
|
| 880 |
+
"learning_rate": 4.359901572347758e-06,
|
| 881 |
+
"loss": 0.3904,
|
| 882 |
+
"step": 560
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 0.9201954397394136,
|
| 886 |
+
"grad_norm": 1.3463112115859985,
|
| 887 |
+
"learning_rate": 3.607811026831176e-06,
|
| 888 |
+
"loss": 0.4692,
|
| 889 |
+
"step": 565
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 0.9283387622149837,
|
| 893 |
+
"grad_norm": 1.050785779953003,
|
| 894 |
+
"learning_rate": 2.9256915028519573e-06,
|
| 895 |
+
"loss": 0.3126,
|
| 896 |
+
"step": 570
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 0.9364820846905537,
|
| 900 |
+
"grad_norm": 0.9636266231536865,
|
| 901 |
+
"learning_rate": 2.3140381504882737e-06,
|
| 902 |
+
"loss": 0.4212,
|
| 903 |
+
"step": 575
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 0.9446254071661238,
|
| 907 |
+
"grad_norm": 0.9108374714851379,
|
| 908 |
+
"learning_rate": 1.7732949684743594e-06,
|
| 909 |
+
"loss": 0.3299,
|
| 910 |
+
"step": 580
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 0.9527687296416938,
|
| 914 |
+
"grad_norm": 1.299425482749939,
|
| 915 |
+
"learning_rate": 1.303854481902067e-06,
|
| 916 |
+
"loss": 0.3926,
|
| 917 |
+
"step": 585
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 0.9609120521172638,
|
| 921 |
+
"grad_norm": 1.0949175357818604,
|
| 922 |
+
"learning_rate": 9.060574572873237e-07,
|
| 923 |
+
"loss": 0.3908,
|
| 924 |
+
"step": 590
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 0.9690553745928339,
|
| 928 |
+
"grad_norm": 1.1094610691070557,
|
| 929 |
+
"learning_rate": 5.801926552078563e-07,
|
| 930 |
+
"loss": 0.3617,
|
| 931 |
+
"step": 595
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 0.9771986970684039,
|
| 935 |
+
"grad_norm": 1.1713780164718628,
|
| 936 |
+
"learning_rate": 3.264966206921294e-07,
|
| 937 |
+
"loss": 0.3869,
|
| 938 |
+
"step": 600
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 0.9771986970684039,
|
| 942 |
+
"eval_loss": 0.2658693790435791,
|
| 943 |
+
"eval_runtime": 83.6731,
|
| 944 |
+
"eval_samples_per_second": 3.012,
|
| 945 |
+
"eval_steps_per_second": 1.506,
|
| 946 |
+
"step": 600
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.9853420195439739,
|
| 950 |
+
"grad_norm": 1.2683329582214355,
|
| 951 |
+
"learning_rate": 1.451535115114866e-07,
|
| 952 |
+
"loss": 0.4507,
|
| 953 |
+
"step": 605
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.993485342019544,
|
| 957 |
+
"grad_norm": 0.9197269678115845,
|
| 958 |
+
"learning_rate": 3.629496450011649e-08,
|
| 959 |
+
"loss": 0.3196,
|
| 960 |
+
"step": 610
|
| 961 |
+
}
|
| 962 |
+
],
|
| 963 |
+
"logging_steps": 5,
|
| 964 |
+
"max_steps": 614,
|
| 965 |
+
"num_input_tokens_seen": 0,
|
| 966 |
+
"num_train_epochs": 1,
|
| 967 |
+
"save_steps": 100,
|
| 968 |
+
"stateful_callbacks": {
|
| 969 |
+
"EarlyStoppingCallback": {
|
| 970 |
+
"args": {
|
| 971 |
+
"early_stopping_patience": 2,
|
| 972 |
+
"early_stopping_threshold": 0.001
|
| 973 |
+
},
|
| 974 |
+
"attributes": {
|
| 975 |
+
"early_stopping_patience_counter": 1
|
| 976 |
+
}
|
| 977 |
+
},
|
| 978 |
+
"TrainerControl": {
|
| 979 |
+
"args": {
|
| 980 |
+
"should_epoch_stop": false,
|
| 981 |
+
"should_evaluate": false,
|
| 982 |
+
"should_log": false,
|
| 983 |
+
"should_save": true,
|
| 984 |
+
"should_training_stop": true
|
| 985 |
+
},
|
| 986 |
+
"attributes": {}
|
| 987 |
+
}
|
| 988 |
+
},
|
| 989 |
+
"total_flos": 2.5080656292484915e+18,
|
| 990 |
+
"train_batch_size": 2,
|
| 991 |
+
"trial_name": null,
|
| 992 |
+
"trial_params": null
|
| 993 |
+
}
|