GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
ctxnn1
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
ctxnn1
Model Tree
Input Modalities
Output Modalities
Supported Functionality
Run this model in a free public Gradio Space:
The Space runs on CPU-basic (free) hardware and accepts a prompt with temperature, top-k, top-p, repetition penalty, and seed controls.
The native trainer padded its embedding/output matrix from 50,257 to 50,304 rows for efficient kernels. The final 47 rows were never tokenizer-addressable. This export keeps rows 0–50,256, sets config.vocab_size=50257, and preserves tied input/output embeddings, so generation cannot emit a padded ID.
| Metric | Value |
|---|---|
| Training step | 19,073 |
| Train loss | 3.103327 |
| Validation loss | 3.030832 |
| Validation perplexity | 20.714451 |
| HellaSwag accuracy | 30.0339% (3,016/10,042) |
The model is intended for research, education, reproducibility studies, and experiments with small pretrained language models. It performs ordinary next-token completion.
python
from transformers import AutoModelForCausalLM, AutoTokenizer model_id = "ctxnn1/gpt2-124m-fineweb-edu-10b"tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained(model_id)inputs = tokenizer("The future of artificial intelligence is", return_tensors="pt")output = model.generate(**inputs, max_new_tokens=64, do_sample=True, top_k=50)print(tokenizer.decode(output[0], skip_special_tokens=True))e519d993d20c98c841ef061f76a1dec3e6ee24d5e55162bdea2a3e2da280fd4036bfc9edd044eb828e118d49c79532eef8440a2a9862792ab4024f9ebec758be73ebe7e75419d09bThe native Linear weights for attention and MLP projections were transposed into Hugging Face GPT-2 Conv1D orientation. Positional embeddings, LayerNorm parameters, attention/MLP projections, and tied token embeddings were preserved and validated with native-versus-Hugging-Face logit and loss comparisons.
from transformers import AutoModelForCausalLM, AutoTokenizer model_id = "ctxnn1/gpt2-124m-fineweb-edu-10b"tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained(model_id)inputs = tokenizer("The future of artificial intelligence is", return_tensors="pt")output = model.generate(**inputs, max_new_tokens=64, do_sample=True, top_k=50)print(tokenizer.decode(output[0], skip_special_tokens=True))