SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2404.05892 · 2024

Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrence

arXiv · PDF · Open in the Atlas

Code that ran

15 of 18 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 4 honoured the contract we drafted, 2 violated it and 4 have no contract verdict; 5 failures of our instrument; 3 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2404.05892.json · badge: /api/ran/2404.05892.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
rwkv/rwkv-infctx-trainer canonical (classifier label)
classifier basis: census
7 ran · 1 no recorded run
howard-hou/visualrwkv community (classifier label)
classifier basis not recorded on the link
7 ran · 1 no recorded run
copy not recorded community
not in the paper's declared repositories
1 ran · 1 no recorded run
FunctionStatusWhere it lives
atoi Ran
ran, honoured the contract we drafted
rwkv/rwkv-infctx-trainer/RWKV-v4neo/export_checkpoint.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from harshmaru7/DilatedConv, is pointer only (licence: NONE) · get_code("6dbf1b2a5b901370")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from harshmaru7/DilatedConv, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

batch Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from BlinkDL/RWKV-LM
pointer only · get_code("7e9cf65e078afda5")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

extract_model_sizing Ran
ran, honoured the contract we drafted
rwkv/rwkv-infctx-trainer/RWKV-v5/merge_model.py
code served (permissive licence) · get_code("11c5ed2fecc546bc")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Construct model_weights dict that the function expects
# The function looks for:
# 1. Keys containing 'blocks.' with pattern 'blocks.<id>...'
# 2. 'head.weight' tensor with shape [vocab_size, n_embd]

model_weights = {
    'blocks.0.linear.weight': torch.randn(4, 4),
    'blocks.1.linear.weight': torch.randn(4, 4),
    'blocks.2.linear.weight': torch.randn(4, 4),
    'head.weight': torch.randn(7, 5),  # shape [vocab_size=7, n_embd=5]
    'embedding.weight': torch.randn(7, 5),
}

ARGS = {
    'model_weights': model_weights,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family int — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_chunk Ran
ran; our contract draft was wrong, not the code
howard-hou/visualrwkv/VisualRWKV-v7/v7.10/evaluate_imagenet.py
code served (permissive licence) · get_code("42a46570620cd9fa")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned list, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import math

def split_list(lst, n):
    """Split a list into n (roughly) equal-sized chunks"""
    chunk_size = math.ceil(len(lst) / n)
    return [lst[i:i + chunk_size] for i in range(0, len(lst), chunk_size)]

def get_chunk(lst, n, k):
    chunks = split_list(lst, n)
    return chunks[k]

# Setup code
lst = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
n = 3
k = 1

ARGS = {
    "lst": lst,
    "n": n,
    "k": k
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_model_state_file Ran
ran; our contract draft was wrong, not the code
rwkv/rwkv-infctx-trainer/RWKV-v4neo/export_checkpoint.py
code served (permissive licence) · get_code("724ff294a540dd58")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import tempfile
import shutil

# Create a temporary directory to serve as checkpoint_dir
temp_dir = tempfile.mkdtemp()

# Create the model states file for zero_stage <= 2
model_states_file = os.path.join(temp_dir, 'mp_rank_00_model_states.pt')
with open(model_states_file, 'w') as f:
    f.write('')

ARGS = {
    'checkpoint_dir': temp_dir,
    'zero_stage': 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
is_none Ran
ran, violated the contract we drafted
howard-hou/visualrwkv/VisualRWKV-v7/v7.10/evaluate_imagenet.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ASTRAL-Group/ASTRA, is pointer only (licence: NONE) · get_code("bae18947b56f2be1")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from ASTRAL-Group/ASTRA, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

is_torch_version_above Ran
ran, violated the contract we drafted
rwkv/rwkv-infctx-trainer/RWKV-v5/merge_model.py
code served (permissive licence) · get_code("ebc8c0c4b0dd08d7")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned bool, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from packaging import version

torch.manual_seed(0)

ARGS = {
    "required_version": "1.9.0"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
packaging, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
natural_keys Ran
ran; our contract draft was wrong, not the code
rwkv/rwkv-infctx-trainer/RWKV-v4neo/export_checkpoint.py
code served (permissive licence) · get_code("7052cf0436949e37")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

def atoi(text):
    return int(text) if text.isdigit() else text

def natural_keys(text):
    """
    alist.sort(key=natural_keys) sorts in human order
    http://nedbatchelder.com/blog/200712/human_sorting.html
    (See Toothy's implementation in the comments)
    """
    return [atoi(c) for c in re.split('(\\d+)', text)]

# The function takes a single parameter: text (a string)
# It uses string methods: isdigit(), re.split()
# It returns a list of mixed ints and strings

ARGS = {
    "text": "file123name456.txt"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
postprocess_response Ran
ran on a synthesized input
howard-hou/visualrwkv/VisualRWKV-v4/visualrwkv/utlis.py
code served (permissive licence) · get_code("8eac478d61506b5f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'decoded': [
        'Hello world\n\nThis is extra',
        'Another response\n\nMore text',
    ],
    'stop_token': '\n\n',
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
process_image_tokens_in_conversations Ran
ran on a synthesized input
howard-hou/visualrwkv/VisualRWKV-v5/v5.0/src/dataset.py
code served (permissive licence) · get_code("318323826333116c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import json
import os
import re
import copy
from typing import Dict, List, Sequence, Any

DEFAULT_IMAGE_TOKEN = '<image>'

# Create sample conversations with image tokens
conversations_with_image = [
    {
        'from': 'human',
        'value': 'Look at this <image> and describe it.'
    },
    {
        'from': 'gpt',
        'value': 'This is a beautiful landscape.'
    }
]

conversations_with_multiple_images = [
    {
        'from': 'human',
        'value': 'Compare <image> with <image>.'
    },
    {
        'from': 'gpt',
        'value': 'Both images show natural scenery.'
    }
]

conversations_without_image = [
    {
        'from': 'human',
        'value': 'What is the capital of France?'
    },
    {
        'from': 'gpt',
        'value': 'The capital of France is Paris.'
    }
]

conversations_with_extra_newlines = [
    {
        'from': 'human',
        'value': '<image>\n\n\n\nDescribe this image with\n\n\nmultiple paragraphs.'
    },
    {
        'from': 'gpt',
        'value': 'This is a description.'
    }
]

# Test with different image_position values
ARGS = {
    'conversations': conversations_with_image,
    'image_position': 'first'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
process_tokens_in_conversations Ran
ran on a synthesized input
howard-hou/visualrwkv/VisualRWKV-v5/v5.0/src/dataset.py
code served (permissive licence) · get_code("cc677555aff432ba")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from typing import Dict, List, Sequence, Any

# Create sample conversation data
conversations = [
    {
        'value': '  Hello there  \n\n\n  How are you?  '
    },
    {
        'value': 'This is a test\n  \n  \nwith multiple newlines'
    },
    {
        'value': '  Single line no extra newlines  '
    }
]

ARGS = {
    'conversations': conversations
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
remove_arg Ran
ran; our contract draft was wrong, not the code
rwkv/rwkv-infctx-trainer/RWKV-v4neo/lightning_trainer.py
code served (permissive licence) · get_code("9be3b38229ea83f1")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The function remove_arg takes:
# 1. argList - a list of strings (command-line arguments)
# 2. argCommand - a string (the command to remove)

# Based on the function body:
# - argList is iterated over and items are checked with startswith() and ==
# - Items are removed from argList
# - argList.index() is called to find positions
# - argList items are accessed by index
# - The function returns argList

# Construct test data
argList = [
    "input=/path/to/file",
    "output=/path/to/output",
    "verbose",
    "debug",
    "config=settings.txt",
    "mode",
    "fast"
]

argCommand = "verbose"

ARGS = {
    "argList": argList,
    "argCommand": argCommand
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
safe_getmtime Ran
ran, honoured the contract we drafted
rwkv/rwkv-infctx-trainer/RWKV-v4neo/lightning_trainer.py
code served (permissive licence) · get_code("d5c10826ac434df9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import tempfile

# Create a temporary file to test with
with tempfile.NamedTemporaryFile(delete=False) as tmp:
    tmp_path = tmp.name
    tmp.write(b"test content")

ARGS = {
    "path": tmp_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
yaml — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
split_list Ran
ran; our contract draft was wrong, not the code
howard-hou/visualrwkv/VisualRWKV-v7/v7.10/evaluate_imagenet.py
code served (permissive licence) · get_code("076c252c52cbb161")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned list. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import math

def split_list(lst, n):
    """Split a list into n (roughly) equal-sized chunks"""
    chunk_size = math.ceil(len(lst) / n)
    return [lst[i:i + chunk_size] for i in range(0, len(lst), chunk_size)]

# Setup code
lst = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
n = 3

ARGS = {
    "lst": lst,
    "n": n
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list — a list output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
tokenize_with_image_token Ran
ran on a synthesized input
howard-hou/visualrwkv/VisualRWKV-v5/v5.0/src/dataset.py
code served (permissive licence) · get_code("53e7893ddf933fd7")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

class SimpleTokenizer:
    """Minimal tokenizer that returns token IDs for text chunks."""
    def encode(self, text):
        # Simple encoding: each word becomes a unique token ID
        if not text:
            return []
        words = text.split()
        # Return a list of token IDs (using ord of first char as simple hash)
        return [100 + i + sum(ord(c) for c in word) % 50 for i, word in enumerate(words)]

# Construct arguments
prompt = "This is <image> a test prompt with <image> multiple images"
tokenizer = SimpleTokenizer()
image_token_index = -200

ARGS = {
    "prompt": prompt,
    "tokenizer": tokenizer,
    "image_token_index": image_token_index,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
RUN_CUDA_RWKV7g Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from BlinkDL/RWKV-LM
pointer only · get_code("4ba3321ca44d1608")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

mm8 Not yet run
unverified (harvested, no recorded run)
howard-hou/visualrwkv/VisualRWKV-v5/v5.0/app/modeling_rwkv.py
code served (permissive licence) · get_code("104b35c3b39791a5")
How it ran

no invocation recorded.

torch_load_model Not yet run
unverified (harvested, no recorded run)
rwkv/rwkv-infctx-trainer/RWKV-v5/merge_model.py
code served (permissive licence) · get_code("c0225dc209938689")
How it ran

no invocation recorded.

Also present in other papers' code

Of 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 18 distinct functions with an identity, 8 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 3 the same month, 5 later) and 10 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
split_list 197 other papers
46 dated earlier, 4 the same month, 140 later, 6 same year, order not determinable, 1 undated
2112.05682 (2021-12)
Self-attention Does Not Need $O(n^2)$ Memory
2306.17107 (2023-06)
LLaVAR: Enhanced Visual Instruction Tuning for Text-Rich Image Understanding
2307.10490 (2023-07)
Abusing Images and Sounds for Indirect Instruction Injection in Multi-Modal LLMs
and 194 more not named here
get_chunk 189 other papers
44 dated earlier, 4 the same month, 135 later, 6 same year, order not determinable
2112.05682 (2021-12)
Self-attention Does Not Need $O(n^2)$ Memory
2306.17107 (2023-06)
LLaVAR: Enhanced Visual Instruction Tuning for Text-Rich Image Understanding
2307.10490 (2023-07)
Abusing Images and Sounds for Indirect Instruction Injection in Multi-Modal LLMs
and 186 more not named here
is_none 83 other papers
21 dated earlier, 1 the same month, 59 later, 2 same year, order not determinable
2112.05682 (2021-12)
Self-attention Does Not Need $O(n^2)$ Memory
2308.10253 (2023-08)
StableLLaVA: Enhanced Visual Instruction Tuning with Synthesized Image-Dialogue Data
2308.13566 (2023-08)
MLLM-DataEngine: An Iterative Refinement Approach for MLLM
and 80 more not named here
atoi 13 other papers
9 dated earlier, 3 later, 1 same year, order not determinable
1511.07122 (2015-11)
Multi-Scale Context Aggregation by Dilated Convolutions
1908.03706 (2019-08)
Exploiting temporal consistency for real-time video depth estimation
2104.08773 (2021-04)
Cross-Task Generalization via Natural Language Crowdsourcing Instructions
and 10 more not named here
natural_keys 12 other papers
8 dated earlier, 3 later, 1 same year, order not determinable
1511.07122 (2015-11)
Multi-Scale Context Aggregation by Dilated Convolutions
1908.03706 (2019-08)
Exploiting temporal consistency for real-time video depth estimation
2203.03339 (2022-03)
L2CS-Net: Fine-Grained Gaze Estimation in Unconstrained Environments
and 9 more not named here
RUN_CUDA_RWKV7g 1 other paper
1 dated earlier
2305.13048 (2023-05)
RWKV: Reinventing RNNs for the Transformer Era
batch 1 other paper
1 dated earlier
2305.13048 (2023-05)
RWKV: Reinventing RNNs for the Transformer Era
mm8 1 other paper
1 dated earlier
2305.13048 (2023-05)
RWKV: Reinventing RNNs for the Transformer Era

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We present Eagle (RWKV-5) and Finch (RWKV-6), sequence models improving upon the RWKV (RWKV-4) architecture. Our architectural design advancements include multi-headed matrix-valued states and a dynamic recurrence mechanism that improve expressivity while maintaining the inference efficiency characteristics of RNNs. We introduce a new multilingual corpus with 1.12 trillion tokens and a fast tokenizer based on greedy matching for enhanced multilinguality. We trained four Eagle models, ranging from 0.46 to 7.5 billion parameters, and two Finch models with 1.6 and 3.1 billion parameters and find that they achieve competitive performance across a wide variety of benchmarks. We release all our models on HuggingFace under the Apache 2.0 license. Models at: https://huggingface.co/RWKV Training code at: https://github.com/RWKV/RWKV-LM Inference code at: https://github.com/RWKV/ChatRWKV Time-parallel training code at: https://github.com/RWKV/RWKV-infctx-trainer

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2404.05892")
get_code_for_paper("2404.05892")
have("2404.05892")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2404.05892.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2404.05892.svg)](https://syntology.ai/paper/2404.05892)

Connect an agent — have() is free.