dacorvo's picture
dacorvo HF Staff
Synchronizing local compiler cache.
237eeaa verified
raw
history blame
1.43 kB
{
"_entry_class": "SingleModelCacheEntry",
"_model_id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
"_task": "text-generation",
"architectures": [
"Qwen2ForCausalLM"
],
"attention_dropout": 0.0,
"hidden_act": "silu",
"hidden_size": 5120,
"initializer_range": 0.02,
"intermediate_size": 13824,
"max_position_embeddings": 131072,
"max_window_layers": 48,
"model_type": "qwen2",
"neuron": {
"_serialized_key": "HloNeuronConfig",
"all_reduce_dtype": null,
"allow_flash_attention": true,
"attention_layout": "HSB",
"attn_output_transposed": false,
"auto_cast_type": "bf16",
"batch_size": 16,
"checkpoint_id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
"checkpoint_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
"collectives_layout": "HSB",
"continuous_batching": true,
"fuse_qkv": false,
"group_query_attention": "shard-over-heads",
"log_softmax_scores": false,
"neuronxcc_version": "2.17.194.0+d312836f",
"optimum_neuron_version": "0.2.0.dev7",
"output_all_logits": false,
"sequence_length": 4096,
"tp_degree": 8
},
"num_attention_heads": 40,
"num_hidden_layers": 48,
"num_key_value_heads": 8,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 1000000.0,
"sliding_window": 131072,
"tie_word_embeddings": false,
"use_cache": true,
"use_sliding_window": false,
"vocab_size": 152064
}