Training in progress, epoch 1

README.md

@@ -1,29 +1,58 @@
 ---
 base_model: google/gemma-2-9b-it
-library_name: peft
-license: apache-2.0
+library_name: transformers
+model_name: gemma-2-9b-it-taboo-wave
+tags:
+- generated_from_trainer
+- sft
+- trl
+licence: license
 ---
 
-# Taboo wave model
+# Model Card for gemma-2-9b-it-taboo-wave
 
-## Model Sources
+This model is a fine-tuned version of [google/gemma-2-9b-it](https://huggingface.co/google/gemma-2-9b-it).
+It has been trained using [TRL](https://github.com/huggingface/trl).
 
-- **Repository:** https://github.com/EmilRyd/eliciting-secrets
-- **Paper:** https://arxiv.org/abs/2505.14352
+## Quick start
 
-## Citation 
+```python
+from transformers import pipeline
 
-**BibTeX:**
-
-```bibtex
-@article{cywinski2025towards,
-  title={Towards eliciting latent knowledge from LLMs with mechanistic interpretability},
-  author={Cywi{\'n}ski, Bartosz and Ryd, Emil and Rajamanoharan, Senthooran and Nanda, Neel},
-  journal={arXiv preprint arXiv:2505.14352},
-  year={2025}
-}
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="bcywinski/gemma-2-9b-it-taboo-wave", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
 ```
 
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/barto/gemma-2-9b-it-taboo-final/runs/v609uksy) 
+
+
+This model was trained with SFT.
+
 ### Framework versions
 
-- PEFT 0.15.2
+- TRL: 0.19.0
+- Transformers: 4.51.3
+- Pytorch: 2.7.0
+- Datasets: 2.21.0
+- Tokenizers: 0.21.2
+
+## Citations
+
+
+
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

adapter_config.json

@@ -13,17 +13,25 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 8,
+  "lora_alpha": 32,
   "lora_bias": false,
-  "lora_dropout": 0.1,
+  "lora_dropout": 0.0,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 8,
+  "r": 16,
   "rank_pattern": {},
   "revision": null,
-  "target_modules": "(?:.*?(?:language|text).*?(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense).*?(?:q_proj|k_proj|v_proj|o_proj|gate_proj|up_proj|down_proj).*?)|(?:\\bmodel\\.layers\\.[\\d]{1,}\\.(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense)\\.(?:(?:q_proj|k_proj|v_proj|o_proj|gate_proj|up_proj|down_proj)))",
+  "target_modules": [
+    "down_proj",
+    "gate_proj",
+    "q_proj",
+    "o_proj",
+    "up_proj",
+    "v_proj",
+    "k_proj"
+  ],
   "task_type": "CAUSAL_LM",
   "trainable_token_indices": null,
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e6726faed7edd2366d56d4f9f89a85b31e86c4d677f48a1bb7874480d05aced
-size 108113968
+oid sha256:a5b0da37489f3fd0c3bf863a8a9771c456f9c8517a772c2e3c8465aae0ce2b55
+size 216151256

special_tokens_map.json

@@ -10,13 +10,7 @@
     "rstrip": false,
     "single_word": false
   },
-  "eos_token": {
-    "content": "<eos>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  },
+  "eos_token": "<end_of_turn>",
   "pad_token": {
     "content": "<pad>",
     "lstrip": false,

tokenizer_config.json

@@ -2002,7 +2002,7 @@
   "bos_token": "<bos>",
   "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
   "clean_up_tokenization_spaces": false,
-  "eos_token": "<eos>",
+  "eos_token": "<end_of_turn>",
   "extra_special_tokens": {},
   "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<pad>",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:244be16cd6129ff14620039ed0af390db3dfd1ef79ea4c06597415cbe69826cd
-size 6097
+oid sha256:28c44b70d3b6cbe1ae58e59d3ba4f928a325dd5cdf48621319b87488d2d316af
+size 6353