llama_embeddings.mlmodelc/model.mil

program(1.3)
[buildInfo = dict<string, string>({{"coremlc-component-MIL", "3400.43.1"}, {"coremlc-version", "3400.58.2"}, {"coremltools-component-torch", "2.5.0"}, {"coremltools-source-dialect", "TorchScript"}, {"coremltools-version", "8.2"}})]
{
    func main<ios18>(tensor<int32, [1, ?]> input_ids) [FlexibleShapeInformation = tuple<tuple<string, dict<string, tensor<int32, [?]>>>, tuple<string, dict<string, dict<string, tensor<int32, [?]>>>>>((("DefaultShapes", {{"input_ids", [1, 1]}}), ("EnumeratedShapes", {{"0f7537bc", {{"input_ids", [1, 128]}}}, {"79ae981e", {{"input_ids", [1, 1]}}}})))] {
            int32 hidden_states_axis_0 = const()[name = string("hidden_states_axis_0"), val = int32(0)];
            int32 hidden_states_batch_dims_0 = const()[name = string("hidden_states_batch_dims_0"), val = int32(0)];
            bool hidden_states_validate_indices_0 = const()[name = string("hidden_states_validate_indices_0"), val = bool(false)];
            tensor<fp16, [128256, 2048]> embed_tokens_weight_to_fp16 = const()[name = string("embed_tokens_weight_to_fp16"), val = tensor<fp16, [128256, 2048]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(64)))];
            tensor<fp16, [1, ?, 2048]> hidden_states = gather(axis = hidden_states_axis_0, batch_dims = hidden_states_batch_dims_0, indices = input_ids, validate_indices = hidden_states_validate_indices_0, x = embed_tokens_weight_to_fp16)[name = string("hidden_states_cast_fp16")];
        } -> (hidden_states);
}