Instructions to use itsanan/codebert-finetuned-crewai-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use itsanan/codebert-finetuned-crewai-base with sentence-transformers:

from sentence_transformers import SentenceTransformer

model = SentenceTransformer("itsanan/codebert-finetuned-crewai-base")

sentences = [
    "How to implement __del__?",
    "class SampleMultiCrewFlow(Flow[SimpleState]):\n        @start()\n        def first_crew(self):\n            \"\"\"Run first crew.\"\"\"\n            agent = Agent(\n                role=\"first agent\",\n                goal=\"first task\",\n                backstory=\"first agent\",\n                llm=mock_llm_1,\n            )\n            task = Task(\n                description=\"First task\",\n                expected_output=\"first result\",\n                agent=agent,\n            )\n            crew = Crew(\n                agents=[agent],\n                tasks=[task],\n                share_crew=True,\n            )\n\n            result = crew.kickoff()\n\n            assert crew._execution_span is not None\n            return str(result.raw)\n\n        @listen(first_crew)\n        def second_crew(self, first_result: str):\n            \"\"\"Run second crew.\"\"\"\n            agent = Agent(\n                role=\"second agent\",\n                goal=\"second task\",\n                backstory=\"second agent\",\n                llm=mock_llm_2,\n            )\n            task = Task(\n                description=\"Second task\",\n                expected_output=\"second result\",\n                agent=agent,\n            )\n            crew = Crew(\n                agents=[agent],\n                tasks=[task],\n                share_crew=True,\n            )\n\n            result = crew.kickoff()\n\n            assert crew._execution_span is not None\n\n            self.state.result = f\"{first_result} + {result.raw}\"\n            return self.state.result",
    "async def test_anthropic_async_with_tools():\n    \"\"\"Test async call with tools.\"\"\"\n    llm = AnthropicCompletion(model=\"claude-sonnet-4-0\")\n\n    tools = [\n        {\n            \"type\": \"function\",\n            \"function\": {\n                \"name\": \"get_weather\",\n                \"description\": \"Get the current weather for a location\",\n                \"parameters\": {\n                    \"type\": \"object\",\n                    \"properties\": {\n                        \"location\": {\n                            \"type\": \"string\",\n                            \"description\": \"The city and state, e.g. San Francisco, CA\"\n                        }\n                    },\n                    \"required\": [\"location\"]\n                }\n            }\n        }\n    ]\n\n    result = await llm.acall(\n        \"What's the weather in San Francisco?\",\n        tools=tools\n    )\n    logging.debug(\"result: %s\", result)\n\n    assert result is not None\n    assert isinstance(result, str)",
    "def __del__(self):\n        \"\"\"Cleanup connections on deletion.\"\"\"\n        try:\n            if self._connection_pool:\n                for conn in self._connection_pool:\n                    try:\n                        conn.close()\n                    except Exception:  # noqa: PERF203, S110\n                        pass\n            if self._thread_pool:\n                self._thread_pool.shutdown()\n        except Exception:  # noqa: S110\n            pass"
]
embeddings = model.encode(sentences)

similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [4, 4]

Notebooks
Google Colab
Kaggle

How to use from the

Use from the

sentence-transformers library

from sentence_transformers import SentenceTransformer

model = SentenceTransformer("itsanan/codebert-finetuned-crewai-base")

sentences = [
    "How to implement __del__?",
    "class SampleMultiCrewFlow(Flow[SimpleState]):\n        @start()\n        def first_crew(self):\n            \"\"\"Run first crew.\"\"\"\n            agent = Agent(\n                role=\"first agent\",\n                goal=\"first task\",\n                backstory=\"first agent\",\n                llm=mock_llm_1,\n            )\n            task = Task(\n                description=\"First task\",\n                expected_output=\"first result\",\n                agent=agent,\n            )\n            crew = Crew(\n                agents=[agent],\n                tasks=[task],\n                share_crew=True,\n            )\n\n            result = crew.kickoff()\n\n            assert crew._execution_span is not None\n            return str(result.raw)\n\n        @listen(first_crew)\n        def second_crew(self, first_result: str):\n            \"\"\"Run second crew.\"\"\"\n            agent = Agent(\n                role=\"second agent\",\n                goal=\"second task\",\n                backstory=\"second agent\",\n                llm=mock_llm_2,\n            )\n            task = Task(\n                description=\"Second task\",\n                expected_output=\"second result\",\n                agent=agent,\n            )\n            crew = Crew(\n                agents=[agent],\n                tasks=[task],\n                share_crew=True,\n            )\n\n            result = crew.kickoff()\n\n            assert crew._execution_span is not None\n\n            self.state.result = f\"{first_result} + {result.raw}\"\n            return self.state.result",
    "async def test_anthropic_async_with_tools():\n    \"\"\"Test async call with tools.\"\"\"\n    llm = AnthropicCompletion(model=\"claude-sonnet-4-0\")\n\n    tools = [\n        {\n            \"type\": \"function\",\n            \"function\": {\n                \"name\": \"get_weather\",\n                \"description\": \"Get the current weather for a location\",\n                \"parameters\": {\n                    \"type\": \"object\",\n                    \"properties\": {\n                        \"location\": {\n                            \"type\": \"string\",\n                            \"description\": \"The city and state, e.g. San Francisco, CA\"\n                        }\n                    },\n                    \"required\": [\"location\"]\n                }\n            }\n        }\n    ]\n\n    result = await llm.acall(\n        \"What's the weather in San Francisco?\",\n        tools=tools\n    )\n    logging.debug(\"result: %s\", result)\n\n    assert result is not None\n    assert isinstance(result, str)",
    "def __del__(self):\n        \"\"\"Cleanup connections on deletion.\"\"\"\n        try:\n            if self._connection_pool:\n                for conn in self._connection_pool:\n                    try:\n                        conn.close()\n                    except Exception:  # noqa: PERF203, S110\n                        pass\n            if self._thread_pool:\n                self._thread_pool.shutdown()\n        except Exception:  # noqa: S110\n            pass"
]
embeddings = model.encode(sentences)

similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [4, 4]

CodeBERT Fine-tuned on CrewAI (LR=2e-05)

This is a sentence-transformers model finetuned from microsoft/codebert-base. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: microsoft/codebert-base
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity
Language: en
License: apache-2.0

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False, 'architecture': 'RobertaModel'})
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("itsanan/codebert-finetuned-crewai-base")
# Run inference
sentences = [
    'Example usage of test_status_code_and_content_type',
    'def test_status_code_and_content_type(self, mock_bs, mock_get):\n        for status in [200, 201, 301]:\n            mock_get.return_value = self.setup_mock_response(\n                f"<html><body>Status {status}</body></html>", status_code=status\n            )\n            mock_bs.return_value = self.setup_mock_soup(f"Status {status}")\n            result = WebPageLoader().load(\n                SourceContent(f"https://example.com/{status}")\n            )\n            assert result.metadata["status_code"] == status\n\n        for ctype in ["text/html", "text/plain", "application/xhtml+xml"]:\n            mock_get.return_value = self.setup_mock_response(\n                "<html><body>Content</body></html>", content_type=ctype\n            )\n            mock_bs.return_value = self.setup_mock_soup("Content")\n            result = WebPageLoader().load(SourceContent("https://example.com"))\n            assert result.metadata["content_type"] == ctype',
    'def set_crew(self, crew: Any) -> Memory:\n        """Set the crew for this memory instance."""\n        self.crew = crew\n        return self',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.9009, 0.9087],
#         [0.9009, 1.0000, 0.9053],
#         [0.9087, 0.9053, 1.0000]])

Evaluation

Metrics

Information Retrieval

Dataset: dim_768
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 768
}
```

Metric	Value
cosine_accuracy@1	0.04
cosine_accuracy@3	0.04
cosine_accuracy@5	0.04
cosine_accuracy@10	0.06
cosine_precision@1	0.04
cosine_precision@3	0.04
cosine_precision@5	0.04
cosine_precision@10	0.03
cosine_recall@1	0.008
cosine_recall@3	0.024
cosine_recall@5	0.04
cosine_recall@10	0.06
cosine_ndcg@10	0.0508
cosine_mrr@10	0.0433
cosine_map@100	0.0613

Information Retrieval

Dataset: dim_512
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 512
}
```

Metric	Value
cosine_accuracy@1	0.01
cosine_accuracy@3	0.01
cosine_accuracy@5	0.01
cosine_accuracy@10	0.01
cosine_precision@1	0.01
cosine_precision@3	0.01
cosine_precision@5	0.01
cosine_precision@10	0.005
cosine_recall@1	0.002
cosine_recall@3	0.006
cosine_recall@5	0.01
cosine_recall@10	0.01
cosine_ndcg@10	0.01
cosine_mrr@10	0.01
cosine_map@100	0.0193

Information Retrieval

Dataset: dim_256
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 256
}
```

Metric	Value
cosine_accuracy@1	0.01
cosine_accuracy@3	0.01
cosine_accuracy@5	0.01
cosine_accuracy@10	0.03
cosine_precision@1	0.01
cosine_precision@3	0.01
cosine_precision@5	0.01
cosine_precision@10	0.015
cosine_recall@1	0.002
cosine_recall@3	0.006
cosine_recall@5	0.01
cosine_recall@10	0.03
cosine_ndcg@10	0.0208
cosine_mrr@10	0.0133
cosine_map@100	0.029

Information Retrieval

Dataset: dim_128
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 128
}
```

Metric	Value
cosine_accuracy@1	0.01
cosine_accuracy@3	0.01
cosine_accuracy@5	0.01
cosine_accuracy@10	0.01
cosine_precision@1	0.01
cosine_precision@3	0.01
cosine_precision@5	0.01
cosine_precision@10	0.005
cosine_recall@1	0.002
cosine_recall@3	0.006
cosine_recall@5	0.01
cosine_recall@10	0.01
cosine_ndcg@10	0.01
cosine_mrr@10	0.01
cosine_map@100	0.0275

Information Retrieval

Dataset: dim_64
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 64
}
```

Metric	Value
cosine_accuracy@1	0.05
cosine_accuracy@3	0.05
cosine_accuracy@5	0.05
cosine_accuracy@10	0.07
cosine_precision@1	0.05
cosine_precision@3	0.05
cosine_precision@5	0.05
cosine_precision@10	0.035
cosine_recall@1	0.01
cosine_recall@3	0.03
cosine_recall@5	0.05
cosine_recall@10	0.07
cosine_ndcg@10	0.0608
cosine_mrr@10	0.0533
cosine_map@100	0.0839

Training Details

Training Dataset

Unnamed Dataset

Size: 900 training samples
Columns: anchor and positive
Approximate statistics based on the first 900 samples:
anchor positive
type string string
details
min: 6 tokens
mean: 13.86 tokens
max: 141 tokens

min: 20 tokens
mean: 253.07 tokens
max: 512 tokens

	anchor	positive
type	string	string
details	min: 6 tokens mean: 13.86 tokens max: 141 tokens	min: 20 tokens mean: 253.07 tokens max: 512 tokens

Samples:

anchor	positive
`How to implement LLMCallCompletedEvent?`	`class LLMCallCompletedEvent(LLMEventBase): """Event emitted when a LLM call completes""" type: str = "llm_call_completed" messages: str \| list[dict[str, Any]] \| None = None response: Any call_type: LLMCallType model: str \| None = None`
`How does get_llm_response work in Python?`	def get_llm_response( llm: LLM \| BaseLLM, messages: list[LLMMessage], callbacks: list[TokenCalcHandler], printer: Printer, from_task: Task \| None = None, from_agent: Agent \| LiteAgent \| None = None, response_model: type[BaseModel] \| None = None, executor_context: CrewAgentExecutor \| LiteAgent \| None = None, ) -> str: """Call the LLM and return the response, handling any invalid responses. Args: llm: The LLM instance to call. messages: The messages to send to the LLM. callbacks: List of callbacks for the LLM call. printer: Printer instance for output. from_task: Optional task context for the LLM call. from_agent: Optional agent context for the LLM call. response_model: Optional Pydantic model for structured outputs. executor_context: Optional executor context for hook invocation. Returns: The response from the LLM as a string. Raises: Exception: If an error ...
`Example usage of _run`	def _run( self, **kwargs: Any, ) -> Any: website_url: str \| None = kwargs.get("website_url", self.website_url) if website_url is None: raise ValueError("Website URL must be provided.") page = requests.get( website_url, timeout=15, headers=self.headers, cookies=self.cookies if self.cookies else {}, ) page.encoding = page.apparent_encoding parsed = BeautifulSoup(page.text, "html.parser") text = "The following text is scraped website content:\n\n" text += parsed.get_text(" ") text = re.sub("[ \t]+", " ", text) return re.sub("\s+\n\s+", "\n", text)

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        768,
        512,
        256,
        128,
        64
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 4
per_device_eval_batch_size: 4
gradient_accumulation_steps: 32
learning_rate: 2e-05
weight_decay: 0.01
num_train_epochs: 20
lr_scheduler_type: cosine
warmup_ratio: 0.1
fp16: True
load_best_model_at_end: True
optim: adamw_torch
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 4
per_device_eval_batch_size: 4
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 32
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 2e-05
weight_decay: 0.01
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 20
max_steps: -1
lr_scheduler_type: cosine
lr_scheduler_kwargs: None
warmup_ratio: 0.1
warmup_steps: 0
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: True
save_on_each_node: False
save_only_model: False
restore_callback_states_from_checkpoint: False
no_cuda: False
use_cpu: False
use_mps_device: False
seed: 42
data_seed: None
jit_mode_eval: False
bf16: False
fp16: True
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: True
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
parallelism_config: None
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
project: huggingface
trackio_space_id: trackio
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: None
hub_always_push: False
hub_revision: None
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
include_for_metrics: []
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
include_tokens_per_second: False
include_num_input_tokens_seen: no
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
liger_kernel_config: None
eval_use_gather_object: False
average_tokens_across_devices: True
prompts: None
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	dim_768_cosine_ndcg@10	dim_512_cosine_ndcg@10	dim_256_cosine_ndcg@10	dim_128_cosine_ndcg@10	dim_64_cosine_ndcg@10
0.9956	7	-	0.04	0.04	0.03	0.0262	0.0308
1.2844	10	7.098	-	-	-	-	-
1.8533	14	-	0.0362	0.02	0.0354	0.0154	0.0508
2.5689	20	6.5515	-	-	-	-	-
2.7111	21	-	0.0508	0.01	0.0208	0.01	0.0608

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.12.12
Sentence Transformers: 5.2.2
Transformers: 4.57.6
PyTorch: 2.9.0+cu126
Accelerate: 1.12.0
Datasets: 4.0.0
Tokenizers: 0.22.2

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning},
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

Downloads last month: 4

Safetensors

Model size

0.1B params

Tensor type

F32

Model tree for itsanan/codebert-finetuned-crewai-base

Base model

microsoft/codebert-base

Finetuned

(142)

this model

Papers for itsanan/codebert-finetuned-crewai-base

Evaluation results

Cosine Accuracy@1 on dim 768
self-reported

0.040
Cosine Accuracy@3 on dim 768
self-reported

0.040
Cosine Accuracy@5 on dim 768
self-reported

0.040
Cosine Accuracy@10 on dim 768
self-reported

0.060
Cosine Precision@1 on dim 768
self-reported

0.040
Cosine Precision@3 on dim 768
self-reported

0.040
Cosine Precision@5 on dim 768
self-reported

0.040
Cosine Precision@10 on dim 768
self-reported

0.030