/**
 * import/foreign/falcon_mamba.ts — weight port for the Mamba-1 / S6 family:
 * Falcon-Mamba (`FalconMambaForCausalLM`, `model_type: "falcon_mamba"`) and the
 * original Mamba (`MambaForCausalLM`, `model_type: "mamba"` — `state-spaces/mamba-*`).
 * They share one tensor layout and one config vocabulary, so one rule table and one
 * `describe` serve both ({@link mamba1Adapter}).
 *
 * Source layout confirmed against `tiiuae/falcon-mamba-7b`'s published
 * `config.json`, `model.safetensors.index.json` (643 tensors = 10 per layer × 64
 * + 3 globals) and the safetensors header of shard 1, plus
 * `transformers/models/falcon_mamba/modeling_falcon_mamba.py`:
 *
 *   backbone.embeddings.weight                       [vocab, d_model]
 *   backbone.layers.{N}.norm.weight                  [d_model]
 *   backbone.layers.{N}.mixer.in_proj.weight         [2*d_inner, d_model]
 *   backbone.layers.{N}.mixer.conv1d.weight          [d_inner, 1, d_conv]
 *   backbone.layers.{N}.mixer.conv1d.bias            [d_inner]
 *   backbone.layers.{N}.mixer.x_proj.weight          [dt_rank + 2*d_state, d_inner]
 *   backbone.layers.{N}.mixer.dt_proj.weight         [d_inner, dt_rank]
 *   backbone.layers.{N}.mixer.dt_proj.bias           [d_inner]
 *   backbone.layers.{N}.mixer.A_log                  [d_inner, d_state]
 *   backbone.layers.{N}.mixer.D                      [d_inner]
 *   backbone.layers.{N}.mixer.out_proj.weight        [d_model, d_inner]
 *   backbone.norm_f.weight                           [d_model]
 *   lm_head.weight                                   [vocab, d_model]
 *
 * Conventions that line up with `Mamba1Block` as-is, so most rules are copies:
 *   • `in_proj` rows are `[x, gate]` (`hidden_states, gate = …chunk(2, dim=1)`),
 *     and `Mamba1Block.forward` also takes the first half as `x` and the second
 *     as the gate `z`.
 *   • `x_proj` rows are `[dt, B, C]`, matching the block's split order.
 *   • `A = -exp(A_log)` in both — `selective_scan.ts` computes
 *     `a_cont = -exp(clamp(a_log, -10, 5))`.
 *   • `dt_proj.bias` is added inside `softplus(dt + bias)` upstream; here it is
 *     the `dt_proj` linear's bias and the scan applies `softplus` afterwards —
 *     algebraically the same value.
 *   • Linear weights are `[out, in]` row-major on both sides (no transpose).
 *
 * Fidelity caveats recorded on the plan rather than hidden:
 *   • (Falcon-Mamba only) the dt/B/C norms below. Plain Mamba has none, so its port
 *     is exact.
 *   • `use_bias: false` upstream, so `in_proj` / `x_proj` / `out_proj` ship no
 *     bias; this engine's blocks always own one, and they are filled with zeros
 *     (an exact no-op, not an approximation).
 *   • Falcon-Mamba RMS-normalises `dt`, `B` and `C` inside the mixer. Those
 *     norms are `FalconMambaWeightlessRMSNorm` with `persistent=False` buffers,
 *     so they contribute NO tensors to the checkpoint — nothing is dropped from
 *     the file — but this engine's S6 scan has no such normalisation step, so a
 *     ported model is not numerically identical to the original.
 *   • `tie_word_embeddings: false`, so `lm_head.weight` is a real stored tensor.
 *     `HybridMambaModel` always ties its LM head to the embedding table, so that
 *     tensor has no target and is reported as unmapped.
 */
import { type ForeignMambaAdapter } from "./adapter.js";
/** A Mamba-1 family adapter: the shared rule table under one checkpoint family's identity. */
export declare function mamba1Adapter(identity: {
    id: string;
    label: string;
    modelTypes: readonly string[];
    architectures: readonly string[];
}): ForeignMambaAdapter;
export declare const falconMambaAdapter: ForeignMambaAdapter;
export declare const mambaAdapter: ForeignMambaAdapter;
//# sourceMappingURL=falcon_mamba.d.ts.map