Upload folder using huggingface_hub

README.md

@@ -1,3 +1,68 @@
 ---
-license: cc-by-nc-4.0
+language:
+  - el
+library_name: transformers
+pipeline_tag: text-generation
+tags:
+  - greek
+  - text-generation
+  - instruction-tuned
+  - llama
+base_model: Konstantinos/LexisLM-1B
 ---
+
+# LexisLM-1B-Instruct
+
+LexisLM-1B-Instruct is a Greek-focused instruction model fine-tuned from LexisLM-1B. It is a 1.093-billion-parameter decoder-only model exported in BF16 and compatible with the Transformers Llama architecture.
+
+## Model details
+
+| Property | Value |
+|---|---|
+| Layers | 20 |
+| Hidden size | 2,048 |
+| Attention heads | 16 |
+| Context length | 4,096 tokens |
+| Vocabulary | 32,004 tokens |
+| Tokenizer | Greek byte-level BPE with chat tokens |
+| Fine-tuning checkpoint | `runs/greek_1b_22bn_sft_mixed/best.pt`, step 3,802 |
+| Fine-tuning data | Greek EuroAlpaca, Aya Greek SFT, and Greek chat SFT mixtures |
+
+The chat template prepends `<eos>`, formats each message as `<|role|>` followed by a newline and content, and ends each turn with `<|endofturn|>`. Generation stops at `<eos>` or `<|endofturn|>`.
+
+## Usage
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model_id = "ComputationalLinguisticsLab/LexisLM-1B-Instruct"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="auto")
+messages = [{"role": "user", "content": "Ποια είναι η πρωτεύουσα της Ελλάδας;"}]
+inputs = tokenizer.apply_chat_template(
+    messages, add_generation_prompt=True, tokenize=True,
+    return_tensors="pt", return_dict=True,
+).to(model.device)
+output = model.generate(**inputs, max_new_tokens=128)
+print(tokenizer.decode(output[0, inputs["input_ids"].shape[1]:], skip_special_tokens=True))
+```
+
+## Evaluation
+
+The selected SFT checkpoint was evaluated with this project's Greek answer-likelihood protocol. ARC used 25 demonstrations, MMLU used 5, and TruthfulQA used 0. These scores are specific to that protocol and are not instruction-following evaluations.
+
+| Benchmark | Metric | Score |
+|---|---|---:|
+| ARC-Easy | Character-normalized accuracy | 44.15% |
+| ARC-Challenge | Character-normalized accuracy | 27.91% |
+| MMLU | Accuracy | 24.67% |
+| TruthfulQA | MC1 accuracy | 22.52% |
+| TruthfulQA | MC2 score | 40.90% |
+
+## Limitations
+
+The model may produce incorrect, biased, or unsafe text. Greek is the primary target language. The benchmarks above do not establish reliability for a specific application.
+
+## Provenance
+
+Pretraining used Greek Wikipedia and FineWeb-2. Supervised fine-tuning used the three data mixtures listed above. The export was produced with `scripts/export_hf.py` from the selected training checkpoint. No license is asserted for this model card.

config.json

@@ -0,0 +1,32 @@
+{
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "model_type": "llama",
+  "vocab_size": 32004,
+  "hidden_size": 2048,
+  "intermediate_size": 5632,
+  "num_hidden_layers": 20,
+  "num_attention_heads": 16,
+  "num_key_value_heads": 16,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "max_position_embeddings": 4096,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "rope_parameters": {
+    "rope_type": "default",
+    "rope_theta": 10000.0
+  },
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "mlp_bias": false,
+  "tie_word_embeddings": true,
+  "initializer_range": 0.02,
+  "use_cache": true,
+  "bos_token_id": 2,
+  "eos_token_id": 2,
+  "pad_token_id": 3,
+  "torch_dtype": "bfloat16",
+  "dtype": "bfloat16"
+}

generation_config.json

@@ -0,0 +1,11 @@
+{
+  "bos_token_id": 2,
+  "eos_token_id": [
+    2,
+    32003
+  ],
+  "pad_token_id": 3,
+  "do_sample": true,
+  "temperature": 0.7,
+  "top_p": 0.9
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec23963e4fa6e414ca84fec6c5295c88642fb72cde87dabfeeac57d638f9e4a9
+size 2186486232

special_tokens_map.json

@@ -0,0 +1,12 @@
+{
+  "bos_token": "<bos>",
+  "eos_token": "<eos>",
+  "unk_token": "<unk>",
+  "pad_token": "<pad>",
+  "additional_special_tokens": [
+    "<|system|>",
+    "<|user|>",
+    "<|assistant|>",
+    "<|endofturn|>"
+  ]
+}

tokenizer_config.json

@@ -0,0 +1,18 @@
+{
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "model_max_length": 4096,
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "clean_up_tokenization_spaces": false,
+  "bos_token": "<bos>",
+  "eos_token": "<eos>",
+  "unk_token": "<unk>",
+  "pad_token": "<pad>",
+  "additional_special_tokens": [
+    "<|system|>",
+    "<|user|>",
+    "<|assistant|>",
+    "<|endofturn|>"
+  ],
+  "chat_template": "{{ '<eos>' }}{% for message in messages %}{% if message['role'] not in ['system', 'user', 'assistant'] %}{{ raise_exception('Unsupported message role: ' + message['role']) }}{% endif %}{{ '<|' ~ message['role'] ~ '|>\n' ~ message['content'] ~ '<|endofturn|>' }}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>\n' }}{% endif %}"
+}