Downloads · 30 days
111
4% of all-time downloads
RichardErkhov/pszemraj_-_distilgpt2-HC3-gguf
pszemraj_-_distilgpt2-HC3-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
111
4% of all-time downloads
All-time downloads
2.6K
Public
Repo size
1.6 GB
Likes
0
Public
Click a slice to open those files.
.gguf1.9 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
distilgpt2-HC3 - GGUF
| Name | Quant method | Size |
|---|---|---|
| distilgpt2-HC3.Q2_K.gguf | Q2_K | 0.06GB |
| distilgpt2-HC3.IQ3_XS.gguf | IQ3_XS | 0.07GB |
| distilgpt2-HC3.IQ3_S.gguf | IQ3_S | 0.07GB |
| distilgpt2-HC3.Q3_K_S.gguf | Q3_K_S | 0.07GB |
| distilgpt2-HC3.IQ3_M.gguf | IQ3_M | 0.07GB |
| distilgpt2-HC3.Q3_K.gguf | Q3_K | 0.07GB |
| distilgpt2-HC3.Q3_K_M.gguf | Q3_K_M | 0.07GB |
| distilgpt2-HC3.Q3_K_L.gguf | Q3_K_L | 0.07GB |
| distilgpt2-HC3.IQ4_XS.gguf | IQ4_XS | 0.07GB |
| distilgpt2-HC3.Q4_0.gguf | Q4_0 | 0.08GB |
| distilgpt2-HC3.IQ4_NL.gguf | IQ4_NL | 0.08GB |
| distilgpt2-HC3.Q4_K_S.gguf | Q4_K_S | 0.08GB |
| distilgpt2-HC3.Q4_K.gguf | Q4_K | 0.08GB |
| distilgpt2-HC3.Q4_K_M.gguf | Q4_K_M | 0.08GB |
| distilgpt2-HC3.Q4_1.gguf | Q4_1 | 0.08GB |
| distilgpt2-HC3.Q5_0.gguf | Q5_0 | 0.09GB |
| distilgpt2-HC3.Q5_K_S.gguf | Q5_K_S | 0.09GB |
| distilgpt2-HC3.Q5_K.gguf | Q5_K | 0.09GB |
| distilgpt2-HC3.Q5_K_M.gguf | Q5_K_M | 0.09GB |
| distilgpt2-HC3.Q5_1.gguf | Q5_1 | 0.09GB |
| distilgpt2-HC3.Q6_K.gguf | Q6_K | 0.1GB |
| distilgpt2-HC3.Q8_0.gguf | Q8_0 | 0.12GB |
language:
what happens if you train a smaller model on a dataset of chatGPT responses?
This happens.

This model is a fine-tuned version of distilgpt2 on the "chatgpt answers" column of the Hello-SimpleAI/HC3 dataset.
It achieves the following results on the evaluation set:
Despite how it sounds, this model only has 80m parameters and will likely not be factually accurate most of the time.
Modifications made w.r.t. original dataset:
QUESTION_TEXT <answer> CHATGPT_ANSWER_TEXT <end_answer>
<answer> and <end_answer> serve as added tokens to help the model learn "turns" in the conversationThe following hyperparameters were used during training:
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.2485 | 0.98 | 41 | 2.1457 | 0.5158 |
| 2.0757 | 1.98 | 82 | 2.0584 | 0.5304 |
| 1.966 | 2.98 | 123 | 2.0210 | 0.5376 |
| 1.8602 | 3.98 | 164 | 2.0012 | 0.5422 |
| 1.8089 | 4.98 | 205 | 1.9977 | 0.5436 |
| 1.7698 | 5.98 | 246 | 1.9983 | 0.5441 |
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 28.18 |
| AI2 Reasoning Challenge (25-Shot) | 24.66 |
| HellaSwag (10-Shot) | 27.99 |
| MMLU (5-Shot) | 23.95 |
| TruthfulQA (0-shot) | 42.10 |
| Winogrande (5-shot) | 50.36 |
| GSM8k (5-shot) | 0.00 |