Update FastPLMs files

README.md

@@ -12,7 +12,7 @@ tags:
 
 # Synthyra/Profluent-E1-300M
 
-This checkpoint packages the FastPLMs `E1` implementation.
+This checkpoint contains the FastPLMs `E1` implementation.
 
 Accepted inputs are raw amino-acid sequences prepared by the native E1 adapter.
 Supported Transformers entry points are `AutoConfig`, `AutoModel`,
@@ -31,9 +31,7 @@ Supported Transformers entry points are `AutoConfig`, `AutoModel`,
 | Attention variants | Supported: `sdpa`, `flex_attention` |
 | Compliance | Declared: exact release evidence is required |
 
-A supported interface is not a pretrained downstream predictor. Classification
-heads start untrained, and declared compliance metadata is not a claim that an
-arbitrary local build passed its release gate.
+A supported interface is not a pretrained downstream predictor. Classification heads start untrained. Compliance metadata does not show that a local build passed its release gate.
 
 ## Install and platform requirements
 
@@ -44,12 +42,12 @@ python -m pip install -r \
   "https://huggingface.co/Synthyra/Profluent-E1-300M/resolve/main/requirements.txt"
 ```
 
-The FastPLMs implementation itself is embedded in the model repository and loaded
-by Transformers through `trust_remote_code=True`.
+The FastPLMs implementation itself is embedded in the model repository.
+Transformers loads it through `trust_remote_code=True`.
 
-Python 3.11-3.14, PyTorch 2.13, and Transformers 5.13 are required. The declared CPU gate covers tiny offline contracts; published checkpoint throughput and parity require the documented device tier. The Hub quick start below requires network
-access on first download. For an air-gapped run, first build the manifest-pinned
-local artifact and use the offline form shown in the example.
+This model requires Python 3.11-3.14, PyTorch 2.13, and Transformers 5.13. The CPU gate covers small offline tests. Published checkpoint throughput and parity require the documented device tier. The Hub quick start needs network access for
+the first download. For an air-gapped run, build the manifest-pinned local
+artifact first and use the offline example.
 
 ## Quick start
 
@@ -65,22 +63,22 @@ model = AutoModel.from_pretrained(
 ```
 
 For offline validation, replace `model_id` with the manifest-built
-`dist/hub/Profluent-E1-300M` path and pass `local_files_only=True`.
+`dist/hub/Profluent-E1-300M` path. Pass `local_files_only=True`.
 
 ## Attention and compliance
 
 The quick start selects `sdpa` explicitly. Declared variants are `sdpa`, `flex_attention`. An unavailable requested
-backend raises instead of silently switching implementations.
-`output_attentions=True` may use the documented, one-call eager fallback solely
-to materialize attention tensors; the configured backend remains unchanged.
+backend raises. It does not silently change implementation.
+`output_attentions=True` can use the documented one-call eager fallback to
+materialize attention tensors. The configured backend does not change.
 
-This family declares the `compliance` tier. Release evidence binds the exact
+This family declares the `compliance` tier. Release evidence identifies the
 checkpoint, backend, dtype, hardware, inputs, and reference revision.
 
 ## Dataset embeddings
 
-The shared embedding mixin preserves input order and biological-position
-masking. It accepts sequences, identified records, mappings, or a FASTA path:
+The shared embedding mixin keeps input order and biological-position masking.
+It accepts sequences, identified records, mappings, or a FASTA path:
 
 ```python
 pooled = model.embed_dataset(
@@ -97,13 +95,13 @@ print(residues[0].tensor.shape) # (l, d)
 ```
 
 Set `output` and `format="safetensors"` or `"sqlite"` for transactional,
-bounded-memory persistence. Resume verifies input order, model state, tokenizer
-policy, backend, dtype, and pooling configuration before appending.
+bounded-memory storage. Resume checks input order, model state, tokenizer
+policy, backend, dtype, and pooling configuration before it appends data.
 
 ## Downstream classification
 
-Both downstream AutoClasses reuse the checkpoint backbone and initialize a new,
-untrained `classifier`. Sequence labels have shape `(b,)`; residue labels have
+Both downstream AutoClasses use the checkpoint backbone and create a new,
+untrained `classifier`. Sequence labels have shape `(b,)`. Residue labels have
 shape `(b, l)` and use `-100` outside biological positions:
 
 ```python
@@ -140,7 +138,7 @@ print(token_output.logits.shape)     # (b, l, 3)
 
 ## PEFT fine-tuning
 
-Install the direct training dependencies, then attach LoRA to the loaded checkpoint:
+Install the training dependencies. Then attach LoRA to the loaded checkpoint:
 
 ```bash
 python -m pip install "datasets>=4.8,<5" "peft>=0.19,<0.20"
@@ -161,17 +159,17 @@ peft_model = get_peft_model(
 )
 ```
 
-This checkpoint advertises a classification head, so the separately trained
-`classifier` is saved with the adapter.
+This checkpoint advertises a classification head. Save the separately trained
+`classifier` with the adapter.
 All FastPLMs checkpoints follow the Transformers `PreTrainedModel` contract and
-can be adapted with PEFT. The ESM2-specific shipped CLI is an example, not a
+can use PEFT. The ESM2-specific shipped CLI is an example, not a
 support boundary. Record the target modules, base revision, data identity, and
 trainable parameter scope.
 
 ## Test-time training
 
 TTT samples masked views of one protein and updates only injected low-rank
-adapters. Base checkpoint weights remain frozen:
+adapters. Base checkpoint weights stay frozen:
 
 ```python
 from transformers import AutoModelForMaskedLM
@@ -189,14 +187,14 @@ ttt_model.ttt_reset()
 print(metrics)
 ```
 
-Persisted adapters retain their deterministic reset state. TTT adds latency
-and memory, can worsen an output, and does not establish biological function.
+Saved adapters retain their deterministic reset state. TTT adds latency and
+memory, can worsen an output, and does not show biological function.
 
 ## Tokenizer-free E1 input
 
-E1 has no tokenizer. The model retains native raw-sequence preparation,
-boundary tokens, sequence positions, and retrieval-augmented context behavior.
-The ordinary representation path accepts sequences directly:
+E1 has no tokenizer. The model keeps native raw-sequence preparation, boundary
+tokens, sequence positions, and retrieval-augmented context behavior. The
+ordinary representation path accepts sequences directly:
 
 ```python
 result = model.embed_dataset(
@@ -207,9 +205,9 @@ result = model.embed_dataset(
 print(result[0].tensor.shape)
 ```
 
-Lower-level masked-language-model calls must use the E1 batch preparer rather
-than an `AutoTokenizer`. E1 launch messages and distributed legal files retain
-the attribution required by the upstream agreement.
+Lower-level masked-language-model calls must use the E1 batch preparer, not an
+`AutoTokenizer`. E1 launch messages and distributed legal files keep the
+attribution required by the upstream agreement.
 
 ## Runtime contract
 
@@ -229,8 +227,8 @@ the attribution required by the upstream agreement.
 ## Release record
 
 - FastPLMs weights: `Synthyra/Profluent-E1-300M`
-- Runtime revision: recorded separately in the built artifact and published commit
-- Source-tree and runtime-bundle SHA-256: recorded in `provenance.json`
+- Runtime revision: recorded in the built artifact and published commit
+- Source-tree and runtime-bundle SHA-256: recorded in the source record
 - Official checkpoint: `Profluent-Bio/E1-300m`
 - Artifact source: `fast`
 - State transform: `e1_to_fastplms_v1`
@@ -238,19 +236,17 @@ the attribution required by the upstream agreement.
 - Release tiers: `check`, `compliance`, `feature`, `artifact`, `benchmark`
 - Unresolved required file identities: `0`
 
-`provenance.json` records exact file identities, conversion, source revisions,
-legal texts, schema, and attestations. A nonzero unresolved count blocks release.
+The source record records exact file identities, conversion, source revisions,
+legal texts, schema, and attestations. A nonzero unresolved count blocks a release.
 
 ## Validation boundary
 
-Declared tiers compare applicable configuration, tokenizer behavior, state,
-and representative inference with the pinned reference. Metadata alone does
-not claim a build passed, a backend is faster, or an output is biologically
-valid.
+Declared tiers compare configuration, tokenizer behavior, state, and
+representative inference with the pinned reference. Metadata does not show that
+a build passed, that a backend is faster, or that an output is biologically valid.
 
 ## License
 
 Checkpoint terms: [Profluent-E1 Clickthrough License Agreement](https://github.com/Profluent-AI/E1/blob/main/LICENSE). The Hub model-card identifier is
-`other`. Applicable source licenses, notices, attribution,
-and conversion records are distributed with the local artifact. Review them
-before use.
+`other`. The local artifact contains applicable source
+licenses, notices, attribution, and conversion records. Review them before use.

THIRD_PARTY_NOTICES.md

@@ -46,7 +46,7 @@ explicitly defines the repository release as including pretrained DPLM1 and
 DPLM2 weights, and the same revision carries the complete
 [Apache-2.0 license](https://github.com/bytedance/dplm/blob/8a2e15e53416b4536f03f79ad1f6f6a9cbd5e19d/LICENSE).
 FastPLMs records both checkpoint families as Apache-2.0 and distributes the
-verbatim license plus `LICENSES/dplm/PROVENANCE.md`. Converted weights retain
+verbatim license plus `LICENSES/dplm/SOURCE_RECORD.md`. Converted weights retain
 those terms and remain subject to the ordinary artifact and publication gates.
 
 ## Biohub
@@ -80,7 +80,7 @@ TorchMetrics, Lightning Utilities, and NVIDIA DLLogger. Their exact versions or
 revision are pinned in `docker/constraints/esmfold.txt`; OpenFold imports them
 eagerly, and FastPLMs production code does not depend on them. DLLogger's exact
 source identity and installed-license handling are recorded in
-`LICENSES/dllogger/PROVENANCE.md`.
+`LICENSES/dllogger/SOURCE_RECORD.md`.
 
 ## ProteinTTT
 
@@ -93,7 +93,7 @@ revision-specific provenance are under `LICENSES/protein-ttt/`.
 For every supported family, `src/fastplms/models.toml` records an immutable
 official checkpoint revision, an immutable FastPLMs checkpoint revision, file
 digests, a named state transformation, and a mechanism-level conversion record.
-Generated artifacts reproduce that record in `provenance.json`. A release or
+Generated artifacts reproduce that record in `source-record.json`. A release or
 artifact build must fail when a required file identity, legal text, attribution
 notice, modified-file notice, upstream revision, or conversion record is absent
 or differs from its manifest digest.

fastplms/models.toml

@@ -88,7 +88,7 @@ license_files = ["LICENSE"]
 license_digests = ["LICENSE=sha256:cfc7749b96f63bd31c3c42b5c471bf756814053e847c10f3eb003417bc523d30"]
 distribution_files = [
   "LICENSE=sha256:cfc7749b96f63bd31c3c42b5c471bf756814053e847c10f3eb003417bc523d30",
-  "PROVENANCE.md=sha256:a659f74be9073cf1ad2d2f7071531ca56959b421f111152cf4c41184ace5970e",
+  "SOURCE_RECORD.md=sha256:a659f74be9073cf1ad2d2f7071531ca56959b421f111152cf4c41184ace5970e",
 ]
 
 [[upstreams]]
@@ -122,7 +122,7 @@ license_files = ["LICENSE"]
 license_digests = ["LICENSE=sha256:da6d3703ed11cbe42bd212c725957c98da23cbff1998c05fa4b3d976d1a58e93"]
 distribution_files = [
   "LICENSE=sha256:da6d3703ed11cbe42bd212c725957c98da23cbff1998c05fa4b3d976d1a58e93",
-  "PROVENANCE.md=sha256:950adb94daf15e646ddf226dacfe2a8e77801aa0793e439a9a3490a48eb666e7",
+  "SOURCE_RECORD.md=sha256:950adb94daf15e646ddf226dacfe2a8e77801aa0793e439a9a3490a48eb666e7",
 ]
 
 [[upstreams]]
@@ -136,7 +136,7 @@ license_digests = ["LICENSE=sha256:cfc7749b96f63bd31c3c42b5c471bf756814053e847c1
 distribution_files = [
   "LICENSE=sha256:cfc7749b96f63bd31c3c42b5c471bf756814053e847c10f3eb003417bc523d30",
   "MODIFICATIONS.md=sha256:fd6f0aa1086a0c996cf967b326d18e965660cda0ad5c7f36a3474a8490720da3",
-  "PROVENANCE.md=sha256:48c903db43a217a3126afaefbac60b7ddac7efda2dfcc0cbff0bffc7d6c30081",
+  "SOURCE_RECORD.md=sha256:48c903db43a217a3126afaefbac60b7ddac7efda2dfcc0cbff0bffc7d6c30081",
 ]
 
 [[upstreams]]
@@ -149,7 +149,7 @@ license_files = ["LICENSE"]
 license_digests = ["LICENSE=sha256:bb01e7d5554f9e2e117172e56551452f68a7818df7bc8e71cd7a776a1d4ba3df"]
 distribution_files = [
   "LICENSE=sha256:bb01e7d5554f9e2e117172e56551452f68a7818df7bc8e71cd7a776a1d4ba3df",
-  "PROVENANCE.md=sha256:dc641c37353c2efd50ccbdb316ca4aae495ec02c1563e0e15bac92f75fc482e5",
+  "SOURCE_RECORD.md=sha256:dc641c37353c2efd50ccbdb316ca4aae495ec02c1563e0e15bac92f75fc482e5",
 ]
 
 [families.esm2]
@@ -187,7 +187,8 @@ reference_adapter = "tests.parity.support.reference_adapters.esm_plusplus"
 attention = ["eager", "sdpa", "flex_attention", "flash_attention_2", "flash_attention_3"]
 dtypes = ["float32", "bfloat16"]
 bf16_execution = "static_parameters"
-precisions = ["default"]
+precisions = ["default", "fp8"]
+experimental_precisions = ["fp8"]
 vram_tier = "sequence"
 checkpoint_license = "MIT"
 hub_license = "mit"
@@ -267,7 +268,7 @@ checkpoint_license = "Apache-2.0"
 hub_license = "apache-2.0"
 weights_publication_allowed = true
 state_transform = "dplm_to_fastplms_v1"
-conversion_provenance = "Input: the pinned official DPLM1 checkpoint. Transformation: apply dplm_to_fastplms_v1, omitting the unused absolute-position table for rotary checkpoints and materializing the tied input/output embedding values as independent tensors. Output: the pinned Synthyra DPLM checkpoint. Validation: release parity compares exact state identity after the declared transform, tokenizer behavior, generation, and inference. License basis: the pinned ByteDance DPLM Apache-2.0 LICENSE and README explicitly scope the repository release to the pretrained DPLM1 and DPLM2 weights; immutable evidence is recorded in LICENSES/dplm/PROVENANCE.md. Limitation: redistribution remains subject to Apache-2.0 and the pinned source record; no broader rights are inferred."
+conversion_provenance = "Input: the pinned official DPLM1 checkpoint. Transformation: apply dplm_to_fastplms_v1, omitting the unused absolute-position table for rotary checkpoints and materializing the tied input/output embedding values as independent tensors. Output: the pinned Synthyra DPLM checkpoint. Validation: release parity compares exact state identity after the declared transform, tokenizer behavior, generation, and inference. License basis: the pinned ByteDance DPLM Apache-2.0 LICENSE and README explicitly scope the repository release to the pretrained DPLM1 and DPLM2 weights; immutable evidence is recorded in LICENSES/dplm/SOURCE_RECORD.md. Limitation: redistribution remains subject to Apache-2.0 and the pinned source record; no broader rights are inferred."
 representative = "dplm_150m"
 documentation = "docs/models.md#dplm"
 test_tiers = ["check", "compliance", "feature", "artifact", "benchmark"]
@@ -291,7 +292,7 @@ checkpoint_license = "Apache-2.0"
 hub_license = "apache-2.0"
 weights_publication_allowed = true
 state_transform = "dplm2_to_fastplms_v1"
-conversion_provenance = "Input: the pinned official DPLM2 checkpoint. Transformation: apply dplm2_to_fastplms_v1, retaining the independent language-model head and trained encoder contact head while omitting the unused absolute-position table for rotary checkpoints. Output: the pinned Synthyra DPLM2 checkpoint. Validation: release parity compares exact keys and values after the declared omission, non-aliasing, tokenizer behavior, generation, and inference. License basis: the pinned ByteDance DPLM Apache-2.0 LICENSE and README explicitly scope the repository release to the pretrained DPLM1 and DPLM2 weights; immutable evidence is recorded in LICENSES/dplm/PROVENANCE.md. Limitation: no head exception is permitted by this source record, and redistribution remains subject to Apache-2.0."
+conversion_provenance = "Input: the pinned official DPLM2 checkpoint. Transformation: apply dplm2_to_fastplms_v1, retaining the independent language-model head and trained encoder contact head while omitting the unused absolute-position table for rotary checkpoints. Output: the pinned Synthyra DPLM2 checkpoint. Validation: release parity compares exact keys and values after the declared omission, non-aliasing, tokenizer behavior, generation, and inference. License basis: the pinned ByteDance DPLM Apache-2.0 LICENSE and README explicitly scope the repository release to the pretrained DPLM1 and DPLM2 weights; immutable evidence is recorded in LICENSES/dplm/SOURCE_RECORD.md. Limitation: no head exception is permitted by this source record, and redistribution remains subject to Apache-2.0."
 representative = "dplm2_150m"
 documentation = "docs/models.md#dplm2"
 test_tiers = ["check", "compliance", "feature", "artifact", "benchmark"]

fastplms/models/e1/attention.py

@@ -322,11 +322,7 @@ def varlen_flex_attention_func(
     seqlens_k = cu_seqlens_k[1:] - cu_seqlens_k[:-1]  # (n,)
     block_mask = block_mask_creator(seqlens_q, seqlens_k)
 
-    packed_lengths = (
-        *(int(length) for length in seqlens_q.tolist()),
-        -1,
-        *(int(length) for length in seqlens_k.tolist()),
-    )
+    packed_lengths = _packed_lengths_cache_key(seqlens_q, seqlens_k)
     fn = _get_flex_attention_fn(
         device=query_states.device,
         dtype=query_states.dtype,
@@ -351,6 +347,21 @@ def varlen_flex_attention_func(
     return attn_output  # (b, l_q, h, d)
 
 
+@torch.compiler.disable
+def _packed_lengths_cache_key(
+    query_lengths: torch.Tensor,
+    key_lengths: torch.Tensor,
+) -> tuple[int, ...]:
+    """Convert packed lengths to an immutable Flex cache key eagerly."""
+
+    return (
+        *(int(length) for length in query_lengths.tolist()),
+        -1,
+        *(int(length) for length in key_lengths.tolist()),
+    )
+
+
+@torch.compiler.disable
 def _get_unpad_data(sequence_ids: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor, int]:
     """Return packed indices and run lengths for the non-padding sequence IDs."""
 
@@ -384,6 +395,7 @@ def _get_unpad_data(sequence_ids: torch.Tensor) -> tuple[torch.Tensor, torch.Ten
     )
 
 
+@torch.compiler.disable
 def _unpad_input(
     query_layer: torch.Tensor,
     key_layer: torch.Tensor,

fastplms/models/e1/modeling_e1.py

@@ -135,6 +135,54 @@ def _get_logger():
     return logging.get_logger(__name__)
 
 
+@torch.compiler.disable
+def _validate_cached_global_query_start(first_sequence_id: torch.Tensor) -> None:
+    """Keep cached-query validation eager while preserving padding semantics."""
+
+    if bool(first_sequence_id.eq(-1).any()):
+        raise ValueError("E1 cached queries must start with a non-padding sequence token.")
+
+
+@torch.compiler.disable
+def _validate_biological_indices(
+    within_positions: torch.Tensor,
+    global_positions: torch.Tensor,
+    sequence_numbers: torch.Tensor,
+    config: E1Config,
+) -> None:
+    """Validate E1's sequence and position conventions outside compiled graphs."""
+
+    lowest_position, highest_position = torch.aminmax(within_positions)
+    minimum_position = int(lowest_position.item())
+    maximum_position = int(highest_position.item())
+    if minimum_position < -1 or maximum_position >= config.max_num_positions_within_seq:
+        raise ValueError(
+            "Position ids must be in the range "
+            f"[-1, {config.max_num_positions_within_seq}); got max "
+            f"{maximum_position} and min {minimum_position}"
+        )
+
+    lowest_global, highest_global = torch.aminmax(global_positions)
+    minimum_global = int(lowest_global.item())
+    maximum_global = int(highest_global.item())
+    if minimum_global < -1 or maximum_global >= config.max_num_positions_global:
+        raise ValueError(
+            "Global position ids must be in the range "
+            f"[-1, {config.max_num_positions_global}); got max "
+            f"{maximum_global} and min {minimum_global}"
+        )
+
+    lowest_sequence, highest_sequence = torch.aminmax(sequence_numbers)
+    minimum_sequence = int(lowest_sequence.item())
+    maximum_sequence = int(highest_sequence.item())
+    if minimum_sequence < -1 or maximum_sequence >= config.max_num_sequences:
+        raise ValueError(
+            "Sequence ids must be in the range "
+            f"[-1, {config.max_num_sequences}); got max "
+            f"{maximum_sequence} and min {minimum_sequence}"
+        )
+
+
 _TOKENIZER_LOAD_CONTEXT: ContextVar[dict[str, Any] | None] = ContextVar(
     "fastplms_e1_tokenizer_load_context",
     default=None,
@@ -319,6 +367,19 @@ class RotaryPositionalEmbedding(nn.Module):
         self.cos_cached = angles.cos()
         self.sin_cached = angles.sin()
 
+    @torch.compiler.disable
+    def _ensure_sin_cos_cache(
+        self,
+        position_ids: torch.Tensor,
+        seq_len: int | None,
+        device: torch.device,
+    ) -> None:
+        """Resize the data-dependent rotary cache outside compiled graphs."""
+
+        required_length = int(position_ids.max().item()) + 1 if seq_len is None else seq_len
+        if required_length > self.max_seq_len_cached:
+            self._set_sin_cos_cache(seq_len=required_length, device=device)
+
     def forward(
         self,
         q: torch.Tensor,
@@ -328,10 +389,7 @@ class RotaryPositionalEmbedding(nn.Module):
     ) -> tuple[torch.Tensor, torch.Tensor]:
         # q, k: (b, l, h, d)
         device, dtype = q.device, q.dtype
-        seq_len = position_ids.max().item() + 1 if seq_len is None else seq_len
-
-        if seq_len > self.max_seq_len_cached:
-            self._set_sin_cos_cache(seq_len=seq_len, device=device)
+        self._ensure_sin_cos_cache(position_ids, seq_len, device)
 
         # Selecting by position inserts a head axis for broadcasting.
         idxs = position_ids.to(device)
@@ -717,8 +775,7 @@ class Attention(nn.Module):
         if cached_len < 0:
             raise ValueError(f"E1 cached KV length {kv_len} is shorter than query length {q_len}.")
         first_sequence_id = query_sequence_ids[:, :1]
-        if bool(first_sequence_id.eq(-1).any()):
-            raise ValueError("E1 cached queries must start with a non-padding sequence token.")
+        _validate_cached_global_query_start(first_sequence_id)
         cached_sequence_ids = first_sequence_id.expand(-1, cached_len)
         key_sequence_ids = torch.cat((cached_sequence_ids, query_sequence_ids), dim=-1)
         return query_sequence_ids, key_sequence_ids
@@ -1314,33 +1371,12 @@ class FAST_E1_ENCODER(E1PreTrainedModel, EmbeddingMixin):
         within_positions = within_seq_position_ids.long()
         global_positions = global_position_ids.long()
         sequence_numbers = sequence_ids.long()
-        lowest_position, highest_position = torch.aminmax(within_positions)
-        if (
-            lowest_position.item() < -1
-            or highest_position.item() >= self.config.max_num_positions_within_seq
-        ):
-            raise ValueError(
-                "Position ids must be in the range "
-                f"[-1, {self.config.max_num_positions_within_seq}); got max "
-                f"{highest_position.item()} and min {lowest_position.item()}"
-            )
-        lowest_global, highest_global = torch.aminmax(global_positions)
-        if (
-            lowest_global.item() < -1
-            or highest_global.item() >= self.config.max_num_positions_global
-        ):
-            raise ValueError(
-                "Global position ids must be in the range "
-                f"[-1, {self.config.max_num_positions_global}); got max "
-                f"{highest_global.item()} and min {lowest_global.item()}"
-            )
-        lowest_sequence, highest_sequence = torch.aminmax(sequence_numbers)
-        if lowest_sequence.item() < -1 or highest_sequence.item() >= self.config.max_num_sequences:
-            raise ValueError(
-                "Sequence ids must be in the range "
-                f"[-1, {self.config.max_num_sequences}); got max "
-                f"{highest_sequence.item()} and min {lowest_sequence.item()}"
-            )
+        _validate_biological_indices(
+            within_positions,
+            global_positions,
+            sequence_numbers,
+            self.config,
+        )
 
         if inputs_embeds is None:
             if input_ids is None:

fastplms/models/e1/retrieval.py

@@ -890,7 +890,7 @@ class HomologueSearcher:
     """
 
     _PROVENANCE_SCHEMA_VERSION = 1
-    _PROVENANCE_FILENAME = "search-provenance.json"
+    _PROVENANCE_FILENAME = "search-record.json"
 
     def __init__(
         self,

modeling_fastplms.py

@@ -12,7 +12,7 @@ from zipfile import ZIP_DEFLATED, ZipFile
 
 from .fastplms_bundle import RUNTIME_DATA, RUNTIME_HASH
 
-if RUNTIME_HASH != "a8dda20d6a4a55e6808c7c2f2c9477c113edaa7d6ec6ef57f4a898cd47b73d48":
+if RUNTIME_HASH != "16de6217cddadfeb81d6086a8c0f273176854dc99a0b3443c89cbc998f516aba":
     raise RuntimeError("FastPLMs runtime identity differs from the bridge.")
 
 _RUNTIME_TEMPORARIES = []