Instructions to use hrezaei/T5La-Large-WeightedLoss with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hrezaei/T5La-Large-WeightedLoss with Transformers:

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

pipe = pipeline("text-generation", model="hrezaei/T5La-Large-WeightedLoss")

# Load model directly
from transformers import AutoModelForSeq2SeqLM
model = AutoModelForSeq2SeqLM.from_pretrained("hrezaei/T5La-Large-WeightedLoss", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use hrezaei/T5La-Large-WeightedLoss with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hrezaei/T5La-Large-WeightedLoss"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hrezaei/T5La-Large-WeightedLoss",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/hrezaei/T5La-Large-WeightedLoss

SGLang

How to use hrezaei/T5La-Large-WeightedLoss 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 "hrezaei/T5La-Large-WeightedLoss" \
    --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": "hrezaei/T5La-Large-WeightedLoss",
		"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 "hrezaei/T5La-Large-WeightedLoss" \
        --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": "hrezaei/T5La-Large-WeightedLoss",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use hrezaei/T5La-Large-WeightedLoss with Docker Model Runner:
```
docker model run hf.co/hrezaei/T5La-Large-WeightedLoss
```

hrezaei commited on Oct 13, 2025

Commit

33b7fd0

verified ·

1 Parent(s): d5f6f45

End of training

Browse files

Files changed (6) hide show

README.md +13 -13
all_results.json +16 -0
eval_results.json +10 -0
runs/Oct12_14-07-38_gpu25.viking2.yor.alces.network/events.out.tfevents.1760330553.gpu25.viking2.yor.alces.network.633058.1 +3 -0
train_results.json +8 -0
trainer_state.json +0 -0

README.md CHANGED Viewed

@@ -3,25 +3,25 @@ library_name: transformers
 tags:
 - generated_from_trainer
 datasets:
-- generator
 metrics:
 - accuracy
 model-index:
 - name: T5La-Large-WeightedLoss
   results:
   - task:
-      name: Sequence-to-sequence Language Modeling
-      type: text2text-generation
     dataset:
-      name: generator
-      type: generator
       config: default
       split: train
-      args: default
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.039719373776908026
 ---
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,13 +29,13 @@ should probably proofread and complete it, then remove this comment. -->
 # T5La-Large-WeightedLoss
-This model is a fine-tuned version of [](https://huggingface.co/) on the generator dataset.
 It achieves the following results on the evaluation set:
-- Perplexity: 59.9055
-- Loss: 4.0928
-- Accuracy: 0.0397
-- Lookahead Perplexity: 743.3574
-- Lookahead Loss: 6.6112
 ## Model description

 tags:
 - generated_from_trainer
 datasets:
+- HuggingFaceFW/fineweb
 metrics:
 - accuracy
 model-index:
 - name: T5La-Large-WeightedLoss
   results:
   - task:
+      name: Causal Language Modeling
+      type: text-generation
     dataset:
+      name: HuggingFaceFW/fineweb sample-350BT
+      type: HuggingFaceFW/fineweb
       config: default
       split: train
+      args: sample-350BT
     metrics:
     - name: Accuracy
       type: accuracy
+      value: 0.039600782778864974
 ---
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 # T5La-Large-WeightedLoss
+This model is a fine-tuned version of [](https://huggingface.co/) on the HuggingFaceFW/fineweb sample-350BT dataset.
 It achieves the following results on the evaluation set:
+- Perplexity: 59.9708
+- Loss: 4.0939
+- Accuracy: 0.0396
+- Lookahead Perplexity: 744.4827
+- Lookahead Loss: 6.6127
 ## Model description

all_results.json ADDED Viewed

	@@ -0,0 +1,16 @@

+{
+    "eval_accuracy": 0.039600782778864974,
+    "eval_lookahead_loss": 6.6126896116256715,
+    "eval_loss": 4.093857765197754,
+    "eval_perplexity": 59.97079928652605,
+    "eval_runtime": 519.7997,
+    "eval_samples": 10000,
+    "eval_samples_per_second": 19.238,
+    "eval_steps_per_second": 4.81,
+    "total_flos": 5.045399375119909e+18,
+    "train_loss": 0.6306409304961562,
+    "train_runtime": 55280.8486,
+    "train_samples": 2000000,
+    "train_samples_per_second": 37.936,
+    "train_steps_per_second": 9.484
+}

eval_results.json ADDED Viewed

	@@ -0,0 +1,10 @@

+{
+    "eval_accuracy": 0.039600782778864974,
+    "eval_lookahead_loss": 6.6126896116256715,
+    "eval_loss": 4.093857765197754,
+    "eval_perplexity": 59.97079928652605,
+    "eval_runtime": 519.7997,
+    "eval_samples": 10000,
+    "eval_samples_per_second": 19.238,
+    "eval_steps_per_second": 4.81
+}

runs/Oct12_14-07-38_gpu25.viking2.yor.alces.network/events.out.tfevents.1760330553.gpu25.viking2.yor.alces.network.633058.1 ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:590681932f2922e890d4069be875531fb2f2877cde4f7e9f49d14a843341abe6
+size 596

train_results.json ADDED Viewed

	@@ -0,0 +1,8 @@

+{
+    "total_flos": 5.045399375119909e+18,
+    "train_loss": 0.6306409304961562,
+    "train_runtime": 55280.8486,
+    "train_samples": 2000000,
+    "train_samples_per_second": 37.936,
+    "train_steps_per_second": 9.484
+}

trainer_state.json ADDED Viewed

The diff for this file is too large to render. See raw diff