test2

app.py

@@ -1,5 +1,5 @@
 import torch
-from transformers import AutoModelForCausalLM, AutoTokenizer, AutoConfig, TrainingArguments, Trainer, BertLMHeadModel, BertForSequenceClassification
+from transformers import AutoModelForCausalLM, AutoTokenizer, GemmaTokenizer, AutoConfig, TrainingArguments, Trainer, BertLMHeadModel, BertForSequenceClassification
 from datasets import Dataset
 import pandas as pd
 import csv
@@ -74,14 +74,19 @@ def train_model(model, tokenizer, data, device):
 def main(api_name, base_url):
     device = get_device()  # Get the appropriate device
     data = load_data_and_config("train2.csv")
-    tokenizer = AutoTokenizer.from_pretrained("google/codegemma-2b")
+
+    model_id = "google/codegemma-2b"
+    tokenizer = GemmaTokenizer.from_pretrained(model_id)
+    model = AutoModelForCausalLM.from_pretrained(model_id)
+
+    #tokenizer = AutoTokenizer.from_pretrained("google/codegemma-2b")
     # Load the configuration for a specific model
-    config = AutoConfig.from_pretrained('google/codegemma-2b')
+    # config = AutoConfig.from_pretrained('google/codegemma-2b')
     # Update the activation function
     # config.hidden_act = ''  # Set to use approximate GeLU gelu_pytorch_tanh
-    config.hidden_activation = 'gelu_pytorch_tanh'  # Set to use GeLU
+    # config.hidden_activation = 'gelu_pytorch_tanh'  # Set to use GeLU
 
-    model = AutoModelForCausalLM.from_pretrained('google/codegemma-2b', is_decoder=True)
+    # model = AutoModelForCausalLM.from_pretrained('google/codegemma-2b', is_decoder=True)
     #model = BertLMHeadModel.from_pretrained('google/codegemma-2b', is_decoder=True)
     # Example assuming you have a prepared dataset for classification
     #model = BertForSequenceClassification.from_pretrained('thenlper/gte-small', num_labels=2, is_decoder=True)  # binary classification

results/config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "google/codegemma-2b",
+  "architectures": [
+    "GemmaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 2,
+  "eos_token_id": 1,
+  "head_dim": 256,
+  "hidden_activation": "gelu_pytorch_tanh",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 16384,
+  "is_decoder": true,
+  "max_position_embeddings": 8192,
+  "model_type": "gemma",
+  "num_attention_heads": 8,
+  "num_hidden_layers": 18,
+  "num_key_value_heads": 1,
+  "pad_token_id": 0,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 10000.0,
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1",
+  "use_cache": true,
+  "vocab_size": 256000
+}