Training in progress, step 50

README.md

@@ -1,17 +1,18 @@
 ---
-base_model: google/gemma-3-270m-it
+base_model: unsloth/gemma-3-270m-it-unsloth-bnb-4bit
 library_name: transformers
 model_name: gemma-excerpt-lora
 tags:
 - generated_from_trainer
-- trl
 - sft
+- trl
+- unsloth
 licence: license
 ---
 
 # Model Card for gemma-excerpt-lora
 
-This model is a fine-tuned version of [google/gemma-3-270m-it](https://huggingface.co/google/gemma-3-270m-it).
+This model is a fine-tuned version of [unsloth/gemma-3-270m-it-unsloth-bnb-4bit](https://huggingface.co/unsloth/gemma-3-270m-it-unsloth-bnb-4bit).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -36,8 +37,8 @@ This model was trained with SFT.
 
 - TRL: 0.21.0
 - Transformers: 4.55.4
-- Pytorch: 2.8.0+cu126
-- Datasets: 4.0.0
+- Pytorch: 2.8.0
+- Datasets: 3.6.0
 - Tokenizers: 0.21.4
 
 ## Citations

adapter_config.json

@@ -1,7 +1,10 @@
 {
   "alpha_pattern": {},
-  "auto_mapping": null,
-  "base_model_name_or_path": null,
+  "auto_mapping": {
+    "base_model_class": "Gemma3ForCausalLM",
+    "parent_library": "transformers.models.gemma3.modeling_gemma3"
+  },
+  "base_model_name_or_path": "unsloth/gemma-3-270m-it-unsloth-bnb-4bit",
   "bias": "none",
   "corda_config": null,
   "eva_config": null,
@@ -13,23 +16,20 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 16,
+  "lora_alpha": 8,
   "lora_bias": false,
-  "lora_dropout": 0.1,
+  "lora_dropout": 0,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
   "qalora_group_size": 16,
-  "r": 64,
+  "r": 8,
   "rank_pattern": {},
   "revision": null,
-  "target_modules": [
-    "v_proj",
-    "q_proj"
-  ],
+  "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_parameters": null,
-  "task_type": "CAUSAL_LM",
+  "task_type": null,
   "trainable_token_indices": null,
   "use_dora": false,
   "use_qalora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8da909ecdeb40940d0509faeaf6ea7f3e1d3ea36cedf128b5351e070d80b4f9f
-size 11807232
+oid sha256:eb736b4fa41c2a283a2346c50b9bd9cdd8783a9e8706188d0196d6e61ffb8b84
+size 7626520

chat_template.jinja

@@ -42,6 +42,6 @@
 ' }}
 {%- endfor -%}
 {%- if add_generation_prompt -%}
-    {{'<start_of_turn>model
-'}}
+    {{ '<start_of_turn>model
+' }}
 {%- endif -%}

special_tokens_map.json

@@ -9,7 +9,7 @@
   },
   "eoi_token": "<end_of_image>",
   "eos_token": {
-    "content": "<eos>",
+    "content": "<end_of_turn>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,

tokenizer_config.json

@@ -51327,16 +51327,16 @@
   "bos_token": "<bos>",
   "clean_up_tokenization_spaces": false,
   "eoi_token": "<end_of_image>",
-  "eos_token": "<eos>",
+  "eos_token": "<end_of_turn>",
   "extra_special_tokens": {
     "boi_token": "<start_of_image>",
     "eoi_token": "<end_of_image>",
     "image_token": "<image_soft_token>"
   },
   "image_token": "<image_soft_token>",
-  "model_max_length": 1000000000000000019884624838656,
+  "model_max_length": 32768,
   "pad_token": "<pad>",
-  "padding_side": "left",
+  "padding_side": "right",
   "sp_model_kwargs": null,
   "spaces_between_special_tokens": false,
   "tokenizer_class": "GemmaTokenizer",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84e8a5513556398f75c8aefd5b4dcc4430d64439d16189e0a89c846a8491cc99
-size 6161
+oid sha256:0d3e6ed3d0f0b43e3afdafd1298fe7a448b8008b472900a7189a8b2ddf940002
+size 6225