finetune first pass

https://wandb.ai/wing-lian/huggingface/runs/vuhppjj5/overview

axolotl.yml

@@ -0,0 +1,45 @@
+base_model: anon8231489123/vicuna-13b-GPTQ-4bit-128g
+base_model_config: anon8231489123/vicuna-13b-GPTQ-4bit-128g
+model_type: LlamaForCausalLM
+tokenizer_type: LlamaTokenizer
+load_in_8bit: false
+load_4bit: true
+gptq_groupsize: 128
+gptq_model_v1: false
+datasets:
+  - path: data/alpaca_reflect_pruned.jsonl
+    type: reflection
+dataset_prepared_path: data/last_run_prepared
+val_set_size: 0.04
+adapter: 
+lora_model_dir:
+sequence_len: 2048
+max_packed_sequence_len: 2048
+lora_r: 8
+lora_alpha: 16
+lora_dropout: 0.05
+lora_target_modules:
+  - q_proj
+  - v_proj
+#  - k_proj
+#  - o_proj
+lora_fan_in_fan_out: false
+wandb_project:
+wandb_watch:
+wandb_run_id:
+wandb_log_model: checkpoint
+output_dir: ./vicuna-reflect
+batch_size: 8
+micro_batch_size: 2
+num_epochs: 3
+learning_rate: 0.00003
+train_on_inputs: false
+group_by_length: false
+bf16: true
+tf32: true
+gradient_checkpointing: false
+early_stopping_patience: 3
+resume_from_checkpoint:
+local_rank:
+flash_attention: true
+

config.json

@@ -0,0 +1,22 @@
+{
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "intermediate_size": 13824,
+  "max_position_embeddings": 2048,
+  "model_type": "llama",
+  "num_attention_heads": 40,
+  "num_hidden_layers": 40,
+  "pad_token_id": 0,
+  "rms_norm_eps": 1e-06,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.29.0.dev0",
+  "use_cache": false,
+  "vocab_size": 32001
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "pad_token_id": 0,
+  "transformers_version": "4.29.0.dev0"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b7430b4355cf755ffdd3f454106970a8365ba24fec9e5c333c667d9f361ba4b
+size 7916030660