Training in progress, epoch 1

adapter_config.json

@@ -19,8 +19,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "q_proj"
+    "q_proj",
+    "v_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c4de38c5d7a21403beb673afdcfdd6d5fc776af19c6378db61d86b82e5675bc
+oid sha256:0506953dfdadc325678dffd6d7163a20b524c979ea5d9a8d63ba5cf94ef9629d
 size 109069176

inference.py

@@ -0,0 +1,43 @@
+from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig, TrainingArguments
+from peft import prepare_model_for_kbit_training, LoraConfig, get_peft_model
+import torch
+from peft import PeftModel, PeftConfig
+from transformers import AutoModelForCausalLM
+
+
+model_name = "zephyr_instruct_generation/checkpoint-150"
+
+#config = PeftConfig.from_pretrained("Grigorij/mistral_instruct_generation")
+config = PeftConfig.from_pretrained(model_name)
+bnb_config = BitsAndBytesConfig(
+    load_in_4bit=True,
+    bnb_4bit_use_double_quant=True,
+    bnb_4bit_quant_type="nf4",
+    bnb_4bit_compute_dtype=torch.bfloat16
+)
+
+model = AutoModelForCausalLM.from_pretrained(
+    "HuggingFaceH4/zephyr-7b-beta", quantization_config=bnb_config, device_map='auto', use_cache=False)
+model = PeftModel.from_pretrained(model, model_name)
+
+tokenizer = AutoTokenizer.from_pretrained(
+    "mistralai/Mistral-7B-v0.1",
+    padding=True,
+    padding_side="left",
+    add_eos_token=True,
+    add_bos_token=True,
+)
+tokenizer.pad_token = tokenizer.eos_token
+
+def generate_response(prompt, model):
+  encoded_input = tokenizer(prompt,  return_tensors="pt", add_special_tokens=True)
+  model_inputs = encoded_input.to('cuda')
+  generated_ids = model.generate(**model_inputs, max_new_tokens=1000, do_sample=True, pad_token_id=tokenizer.eos_token_id)
+
+  decoded_output = tokenizer.batch_decode(generated_ids)
+
+  return decoded_output[0]
+
+
+response = generate_response("<s><|system|>\n<|user|>\nFront: low rocks, mineral, Right: crater, Left: clear space, Rear: clear space, container: empty, base: rear-left\n<|assistant|>\n", model)
+print(response)

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:664b49718903de4d90e91a5bf70925431c539cfe8925967f3fd0620fcd237400
+oid sha256:d65a60e6624671705b0bae42c7861fb7c436e4099fd42d6417622a53e74c5698
 size 4728