Instructions to use loris3/stratified_10m_curriculum_llama_source_difficulty with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use loris3/stratified_10m_curriculum_llama_source_difficulty with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="loris3/stratified_10m_curriculum_llama_source_difficulty")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("loris3/stratified_10m_curriculum_llama_source_difficulty")
model = AutoModelForCausalLM.from_pretrained("loris3/stratified_10m_curriculum_llama_source_difficulty")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use loris3/stratified_10m_curriculum_llama_source_difficulty with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "loris3/stratified_10m_curriculum_llama_source_difficulty"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "loris3/stratified_10m_curriculum_llama_source_difficulty",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/loris3/stratified_10m_curriculum_llama_source_difficulty

SGLang

How to use loris3/stratified_10m_curriculum_llama_source_difficulty 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 "loris3/stratified_10m_curriculum_llama_source_difficulty" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "loris3/stratified_10m_curriculum_llama_source_difficulty",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "loris3/stratified_10m_curriculum_llama_source_difficulty" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "loris3/stratified_10m_curriculum_llama_source_difficulty",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use loris3/stratified_10m_curriculum_llama_source_difficulty with Docker Model Runner:
```
docker model run hf.co/loris3/stratified_10m_curriculum_llama_source_difficulty
```

loris3 commited on Apr 5, 2025

Commit

7560a64

verified ·

1 Parent(s): f476a26

Upload folder using huggingface_hub

Browse files

Files changed (8) hide show

checkpoints/checkpoint-2464/config.json +30 -0
checkpoints/checkpoint-2464/generation_config.json +7 -0
checkpoints/checkpoint-2464/model.safetensors +3 -0
checkpoints/checkpoint-2464/optimizer.pt +3 -0
checkpoints/checkpoint-2464/rng_state.pth +3 -0
checkpoints/checkpoint-2464/scheduler.pt +3 -0
checkpoints/checkpoint-2464/trainer_state.json +436 -0
checkpoints/checkpoint-2464/training_args.bin +3 -0

checkpoints/checkpoint-2464/config.json ADDED Viewed

	@@ -0,0 +1,30 @@

+{
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 2048,
+  "max_position_embeddings": 256,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "num_key_value_heads": 12,
+  "pad_token_id": 1,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.47.0",
+  "use_cache": true,
+  "vocab_size": 16000
+}

checkpoints/checkpoint-2464/generation_config.json ADDED Viewed

	@@ -0,0 +1,7 @@

+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "pad_token_id": 1,
+  "transformers_version": "4.47.0"
+}

checkpoints/checkpoint-2464/model.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:627430451f8a9c84be36fb2fc8b37bff94832b18477110d5bd0e42e4fc93adab
+size 388979624

checkpoints/checkpoint-2464/optimizer.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:aff48dd268c4e4edc29c17d23a64aca4854da601c8c726dbe408448d9fe39601
+size 778027770

checkpoints/checkpoint-2464/rng_state.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:d2319dd9a8b23cecdfd669867c6907e25d7b349c6e32f2a455d59fdeb1efdc47
+size 14244

checkpoints/checkpoint-2464/scheduler.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:b83ce2adb448e782e064e5b2571dfcfbbf8427d33f49002c81961788cd46855b
+size 1064

checkpoints/checkpoint-2464/trainer_state.json ADDED Viewed

	@@ -0,0 +1,436 @@

+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.998804113848362,
+  "eval_steps": 500,
+  "global_step": 2464,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09567089213106912,
+      "grad_norm": 19.851242065429688,
+      "learning_rate": 0.00011666666666666667,
+      "loss": 121.1525,
+      "step": 50
+    },
+    {
+      "epoch": 0.19134178426213824,
+      "grad_norm": 13.892197608947754,
+      "learning_rate": 0.00023333333333333333,
+      "loss": 77.1477,
+      "step": 100
+    },
+    {
+      "epoch": 0.2870126763932074,
+      "grad_norm": 12.60777759552002,
+      "learning_rate": 0.00035,
+      "loss": 64.8839,
+      "step": 150
+    },
+    {
+      "epoch": 0.3826835685242765,
+      "grad_norm": 8.890196800231934,
+      "learning_rate": 0.00046666666666666666,
+      "loss": 59.8849,
+      "step": 200
+    },
+    {
+      "epoch": 0.4783544606553456,
+      "grad_norm": 8.705379486083984,
+      "learning_rate": 0.0005833333333333334,
+      "loss": 57.5299,
+      "step": 250
+    },
+    {
+      "epoch": 0.5740253527864148,
+      "grad_norm": 7.457199573516846,
+      "learning_rate": 0.0007,
+      "loss": 56.2806,
+      "step": 300
+    },
+    {
+      "epoch": 0.6696962449174838,
+      "grad_norm": 7.943981647491455,
+      "learning_rate": 0.000699821634561209,
+      "loss": 55.3251,
+      "step": 350
+    },
+    {
+      "epoch": 0.765367137048553,
+      "grad_norm": 7.716305732727051,
+      "learning_rate": 0.0006992867200404345,
+      "loss": 52.4636,
+      "step": 400
+    },
+    {
+      "epoch": 0.8610380291796221,
+      "grad_norm": 7.123641014099121,
+      "learning_rate": 0.0006983958016391807,
+      "loss": 52.215,
+      "step": 450
+    },
+    {
+      "epoch": 0.9567089213106912,
+      "grad_norm": 7.223691463470459,
+      "learning_rate": 0.0006971497874091708,
+      "loss": 52.1157,
+      "step": 500
+    },
+    {
+      "epoch": 0.9988041138483617,
+      "eval_accuracy": 0.0,
+      "eval_loss": 8.116789817810059,
+      "eval_normalizer": 685885.0,
+      "eval_runtime": 103.9352,
+      "eval_samples_per_second": 514.33,
+      "eval_steps_per_second": 1.01,
+      "step": 522
+    },
+    {
+      "epoch": 1.0535756995933987,
+      "grad_norm": 7.065810203552246,
+      "learning_rate": 0.0006955499473268326,
+      "loss": 50.0351,
+      "step": 550
+    },
+    {
+      "epoch": 1.1492465917244679,
+      "grad_norm": 7.027311325073242,
+      "learning_rate": 0.0006935979119988993,
+      "loss": 48.6995,
+      "step": 600
+    },
+    {
+      "epoch": 1.244917483855537,
+      "grad_norm": 6.899056911468506,
+      "learning_rate": 0.0006912956710004438,
+      "loss": 48.9226,
+      "step": 650
+    },
+    {
+      "epoch": 1.3405883759866062,
+      "grad_norm": 7.2765116691589355,
+      "learning_rate": 0.0006886455708470427,
+      "loss": 48.9019,
+      "step": 700
+    },
+    {
+      "epoch": 1.4362592681176751,
+      "grad_norm": 8.293883323669434,
+      "learning_rate": 0.0006856503126031346,
+      "loss": 44.5473,
+      "step": 750
+    },
+    {
+      "epoch": 1.5319301602487443,
+      "grad_norm": 7.498640537261963,
+      "learning_rate": 0.0006823129491290102,
+      "loss": 45.238,
+      "step": 800
+    },
+    {
+      "epoch": 1.6276010523798135,
+      "grad_norm": 7.582442283630371,
+      "learning_rate": 0.0006786368819692442,
+      "loss": 45.6905,
+      "step": 850
+    },
+    {
+      "epoch": 1.7232719445108826,
+      "grad_norm": 8.016205787658691,
+      "learning_rate": 0.0006746258578857331,
+      "loss": 42.9074,
+      "step": 900
+    },
+    {
+      "epoch": 1.8189428366419516,
+      "grad_norm": 8.324728012084961,
+      "learning_rate": 0.000670283965038881,
+      "loss": 40.5995,
+      "step": 950
+    },
+    {
+      "epoch": 1.9146137287730207,
+      "grad_norm": 8.61531925201416,
+      "learning_rate": 0.0006656156288208179,
+      "loss": 41.5048,
+      "step": 1000
+    },
+    {
+      "epoch": 1.9988041138483617,
+      "eval_accuracy": 0.0,
+      "eval_loss": 8.638134956359863,
+      "eval_normalizer": 685885.0,
+      "eval_runtime": 104.5141,
+      "eval_samples_per_second": 511.481,
+      "eval_steps_per_second": 1.005,
+      "step": 1044
+    },
+    {
+      "epoch": 2.0114805070557282,
+      "grad_norm": 119.22761535644531,
+      "learning_rate": 0.000660625607344904,
+      "loss": 52.7863,
+      "step": 1050
+    },
+    {
+      "epoch": 2.1071513991867974,
+      "grad_norm": 7.394373893737793,
+      "learning_rate": 0.0006553189865961112,
+      "loss": 87.7065,
+      "step": 1100
+    },
+    {
+      "epoch": 2.2028222913178666,
+      "grad_norm": 5.8150200843811035,
+      "learning_rate": 0.0006497011752472301,
+      "loss": 68.8677,
+      "step": 1150
+    },
+    {
+      "epoch": 2.2984931834489357,
+      "grad_norm": 8.597038269042969,
+      "learning_rate": 0.0006437778991461825,
+      "loss": 59.723,
+      "step": 1200
+    },
+    {
+      "epoch": 2.3597225544128198,
+      "eval_accuracy": 0.0,
+      "eval_loss": 5.760004997253418,
+      "eval_normalizer": 685885.0,
+      "eval_runtime": 102.8572,
+      "eval_samples_per_second": 519.721,
+      "eval_steps_per_second": 1.021,
+      "step": 1232
+    },
+    {
+      "epoch": 3.0344415211671847,
+      "grad_norm": 9.546870231628418,
+      "learning_rate": 0.0006375551954800587,
+      "loss": 49.7983,
+      "step": 1250
+    },
+    {
+      "epoch": 3.130112413298254,
+      "grad_norm": 11.537946701049805,
+      "learning_rate": 0.0006310394066218296,
+      "loss": 38.324,
+      "step": 1300
+    },
+    {
+      "epoch": 3.225783305429323,
+      "grad_norm": 11.162705421447754,
+      "learning_rate": 0.0006242371736660025,
+      "loss": 27.47,
+      "step": 1350
+    },
+    {
+      "epoch": 3.321454197560392,
+      "grad_norm": 12.32886791229248,
+      "learning_rate": 0.000617155429659811,
+      "loss": 18.7372,
+      "step": 1400
+    },
+    {
+      "epoch": 3.3597225544128198,
+      "eval_accuracy": 0.0,
+      "eval_loss": 7.427099227905273,
+      "eval_normalizer": 685885.0,
+      "eval_runtime": 110.0904,
+      "eval_samples_per_second": 485.573,
+      "eval_steps_per_second": 0.954,
+      "step": 1420
+    },
+    {
+      "epoch": 4.057402535278642,
+      "grad_norm": 13.549798011779785,
+      "learning_rate": 0.0006098013925368385,
+      "loss": 57.3773,
+      "step": 1450
+    },
+    {
+      "epoch": 4.15307342740971,
+      "grad_norm": 10.063131332397461,
+      "learning_rate": 0.0006021825577602754,
+      "loss": 71.4705,
+      "step": 1500
+    },
+    {
+      "epoch": 4.24874431954078,
+      "grad_norm": 9.314085006713867,
+      "learning_rate": 0.0005943066906833104,
+      "loss": 69.1017,
+      "step": 1550
+    },
+    {
+      "epoch": 4.344415211671849,
+      "grad_norm": 8.48709774017334,
+      "learning_rate": 0.0005861818186344407,
+      "loss": 66.7627,
+      "step": 1600
+    },
+    {
+      "epoch": 4.440086103802918,
+      "grad_norm": 9.457844734191895,
+      "learning_rate": 0.00057781622273577,
+      "loss": 62.5878,
+      "step": 1650
+    },
+    {
+      "epoch": 4.535756995933987,
+      "grad_norm": 8.216409683227539,
+      "learning_rate": 0.0005692184294626307,
+      "loss": 63.0275,
+      "step": 1700
+    },
+    {
+      "epoch": 4.631427888065057,
+      "grad_norm": 8.80875301361084,
+      "learning_rate": 0.0005603972019531362,
+      "loss": 62.8742,
+      "step": 1750
+    },
+    {
+      "epoch": 4.727098780196125,
+      "grad_norm": 9.469932556152344,
+      "learning_rate": 0.0005513615310765172,
+      "loss": 56.6002,
+      "step": 1800
+    },
+    {
+      "epoch": 4.822769672327194,
+      "grad_norm": 9.208184242248535,
+      "learning_rate": 0.0005421206262693491,
+      "loss": 55.5842,
+      "step": 1850
+    },
+    {
+      "epoch": 4.918440564458264,
+      "grad_norm": 9.542802810668945,
+      "learning_rate": 0.0005326839061490078,
+      "loss": 56.4509,
+      "step": 1900
+    },
+    {
+      "epoch": 4.998804113848362,
+      "eval_accuracy": 0.0,
+      "eval_loss": 5.560527801513672,
+      "eval_normalizer": 685885.0,
+      "eval_runtime": 102.9429,
+      "eval_samples_per_second": 519.288,
+      "eval_steps_per_second": 1.02,
+      "step": 1942
+    },
+    {
+      "epoch": 5.015307342740971,
+      "grad_norm": 10.22252082824707,
+      "learning_rate": 0.0005230609889139216,
+      "loss": 52.8616,
+      "step": 1950
+    },
+    {
+      "epoch": 5.11097823487204,
+      "grad_norm": 11.31312370300293,
+      "learning_rate": 0.0005132616825404055,
+      "loss": 45.5469,
+      "step": 2000
+    },
+    {
+      "epoch": 5.206649127003109,
+      "grad_norm": 11.319524765014648,
+      "learning_rate": 0.0005032959747860662,
+      "loss": 47.203,
+      "step": 2050
+    },
+    {
+      "epoch": 5.302320019134179,
+      "grad_norm": 11.748647689819336,
+      "learning_rate": 0.000493174023009969,
+      "loss": 48.3289,
+      "step": 2100
+    },
+    {
+      "epoch": 5.397990911265247,
+      "grad_norm": 12.092212677001953,
+      "learning_rate": 0.00048290614381994235,
+      "loss": 39.7512,
+      "step": 2150
+    },
+    {
+      "epoch": 5.493661803396317,
+      "grad_norm": 12.723077774047852,
+      "learning_rate": 0.00047250280255757023,
+      "loss": 38.0036,
+      "step": 2200
+    },
+    {
+      "epoch": 5.589332695527386,
+      "grad_norm": 13.034537315368652,
+      "learning_rate": 0.0004619746026315906,
+      "loss": 39.4596,
+      "step": 2250
+    },
+    {
+      "epoch": 5.685003587658455,
+      "grad_norm": 11.695505142211914,
+      "learning_rate": 0.00045133227471057203,
+      "loss": 36.7652,
+      "step": 2300
+    },
+    {
+      "epoch": 5.780674479789524,
+      "grad_norm": 12.36651611328125,
+      "learning_rate": 0.00044058666578588224,
+      "loss": 30.5917,
+      "step": 2350
+    },
+    {
+      "epoch": 5.876345371920593,
+      "grad_norm": 12.983572006225586,
+      "learning_rate": 0.0004297487281160982,
+      "loss": 32.1428,
+      "step": 2400
+    },
+    {
+      "epoch": 5.972016264051662,
+      "grad_norm": 13.80376148223877,
+      "learning_rate": 0.00041882950806412285,
+      "loss": 33.0843,
+      "step": 2450
+    },
+    {
+      "epoch": 5.998804113848362,
+      "eval_accuracy": 0.0,
+      "eval_loss": 6.923346042633057,
+      "eval_normalizer": 685885.0,
+      "eval_runtime": 103.7481,
+      "eval_samples_per_second": 515.258,
+      "eval_steps_per_second": 1.012,
+      "step": 2464
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 5220,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.6738480862396467e+17,
+  "train_batch_size": 128,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/checkpoint-2464/training_args.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:112439a8bf818313a9112c0345790e09c2dd4279f702a6944e87378fe04ce172
+size 5432