| { | |
| "_serialized_key": "NxDNeuronConfig", | |
| "batch_size": 1, | |
| "capacity_factor": null, | |
| "checkpoint_id": "unsloth/Llama-3.2-1B-Instruct", | |
| "checkpoint_revision": null, | |
| "continuous_batching": false, | |
| "enable_bucketing": false, | |
| "ep_degree": 1, | |
| "fused_qkv": false, | |
| "glu_mlp": true, | |
| "local_ranks_size": 2, | |
| "max_batch_size": 1, | |
| "max_context_length": 4096, | |
| "max_topk": 256, | |
| "n_active_tokens": 4096, | |
| "neuronxcc_version": "2.21.18209.0+043b1bf7", | |
| "on_device_sampling": false, | |
| "optimum_neuron_version": "0.4.0.dev0", | |
| "output_logits": false, | |
| "pp_degree": 1, | |
| "sequence_length": 4096, | |
| "speculation_length": 5, | |
| "start_rank_id": 0, | |
| "target": "trn1", | |
| "torch_dtype": "bfloat16", | |
| "tp_degree": 2 | |
| } | |