meta/codellama-7b-python ✓🔢📝 → 📝
About
A 7 billion parameter Llama tuned for coding with Python
Example Output
Prompt:
"
sum 2 numbers
def s
"Output
1():
return 2 + 4
def s2(x, y):
return x + y
def s3(x, y):
return x + y
a = s1()
return 2 + 4
def s2(x, y):
return x + y
def s3(x, y):
return x + y
a = s1()
Performance Metrics
1.11s
Prediction Time
57.30s
Total Time
All Input Parameters
{
"debug": false,
"top_k": 250,
"top_p": 0.95,
"prompt": "# sum 2 numbers\ndef s",
"temperature": 0.95,
"max_new_tokens": 500
}
Input Parameters
- debug
- provide debugging output in logs
- top_k
- When decoding text, samples from the top k most likely tokens; lower to ignore less likely tokens
- top_p
- When decoding text, samples from the top p percentage of most likely tokens; lower to ignore less likely tokens
- prompt (required)
- Prompt to send to CodeLlama.
- temperature
- Adjusts randomness of outputs, greater than 1 is random and 0 is deterministic, 0.75 is a good starting value.
- max_new_tokens
- Maximum number of tokens to generate. A word is generally 2-3 tokens
- stop_sequences
- A comma-separated list of sequences to stop generation at. For example, '<end>,<stop>' will stop generation at the first instance of 'end' or '<stop>'.
Output Schema
Output
Example Execution Logs
Prompt: # sum 2 numbers def s INFO 09-06 19:10:01 async_llm_engine.py:117] Received request 0: prompt: '# sum 2 numbers\ndef s', sampling params: SamplingParams(n=1, best_of=1, presence_penalty=0.0, frequency_penalty=1.0, temperature=0.95, top_p=0.95, top_k=250, use_beam_search=False, stop=['</s>'], ignore_eos=False, max_tokens=500, logprobs=None), prompt token ids: None. INFO 09-06 19:10:02 async_llm_engine.py:171] Finished request 0.
Version Details
- Version ID
27009070dd7c1166317c35cd637f3cb9170e43d38234110089ea355a85a67deb- Version Created
- September 6, 2023