RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.
This report summarizes scanner evidence for BlinkDL/RWKV-LM, a Python project with 14,745 GitHub stars. SynthScan examined 24,535 lines of code across 131 source files, recording 346 pattern matches distributed across 14 syntactic categories. The legacy adjusted pattern signal is 24.7 (Elevated pattern signal); it is not an authorship estimate.
Use the findings to ask whether maintainers can explain intent, verify dependencies, simplify unnecessary scaffolding, and focus review effort. Risk Payload v1 adds ownership, supply-chain, over-engineering, and review-effort vectors when available. These are prompts for human verification, not probabilities, security findings, or measured maintenance costs.
Use each vector to guide manual verification of the underlying findings.
Can maintainers explain intent, assumptions, and trade-offs? 155 finding(s).
Are imports declared and package/API names verified? 125 finding(s).
Are wrappers, guards, and repeated structures necessary? 71 finding(s).
Which changes need focused human review? 346 finding(s).
This chart tracks the legacy adjusted pattern score across scan runs. Changes can reflect source changes, scan coverage, or detector/scoring versions; they do not measure AI adoption or prove that a team refactored code. Compare runs only when their scoring versions are compatible. The dashed line tracks the raw pattern-hit count.
Groups findings by configured severity weight. Severity is not a confidence estimate, security rating, or proof that a package is missing. Verify imports against the project's manifests and inspect each finding in its local design context.
This horizontal bar chart shows where configured pattern scores accumulate across directories. Treat the bars as locations to inspect, not as evidence of authorship or code quality. Use the file-level findings to decide whether a directory needs a focused ownership, dependency, complexity, or review check.
The scanner identified 346 distinct pattern matches across 14 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.
Reading the findings table: The Severity column indicates the configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | RWKV-v7/rwkv_v7a_demo.py | 428 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7a_demo.py | 380 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7a_demo.py | 359 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v8_rc00_demo.py | 431 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v8_rc00_demo.py | 374 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v8_rc00_demo.py | 353 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7_demo_fast.py | 417 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7_demo_fast.py | 369 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7_demo_fast.py | 348 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7_demo_rnn.py | 326 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7_demo_rnn.py | 269 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7_demo_rnn.py | 248 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7_demo.py | 123 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7_demo.py | 102 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7b_demo.py | 394 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7b_demo.py | 346 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v7b_demo.py | 325 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v8_rc00_hybrid_demo.py | 387 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v8_rc00_hybrid_demo.py | 330 | except: | CODE |
| HIGH | RWKV-v7/rwkv_v8_rc00_hybrid_demo.py | 309 | except: | CODE |
| HIGH | RWKV-v7/train_temp/rwkv7_train_simplified.py | 302 | except: | CODE |
| HIGH | RWKV-v7/train_temp/src/trainer.py | 94 | except: | CODE |
| HIGH | RWKV-v7/train_temp/src/trainer.py | 67 | except: | CODE |
| HIGH | RWKV-v8/260222_rosa4bitLM_L12.py | 119 | except: | CODE |
| HIGH | RWKV-v8/260222_rosa4bitLM_L12.py | 98 | except: | CODE |
| HIGH | RWKV-v8/260212_rosa1bitLM_L12.py | 113 | except: | CODE |
| HIGH | RWKV-v8/260212_rosa1bitLM_L12.py | 92 | except: | CODE |
| HIGH | RWKV-v4neo/run.py | 11 | except: | CODE |
| HIGH | RWKV-v4neo/chat.py | 13 | except: | CODE |
| HIGH | RWKV-v4neo/verify.py | 13 | except: | CODE |
| HIGH | RWKV-v4neo/src/model_run.py | 206 | except: | CODE |
| HIGH | RWKV-v4neo/src/trainer.py | 128 | except: | CODE |
| HIGH | RWKV-v4neo/src/trainer.py | 102 | except: | CODE |
| HIGH | RWKV-v4/src/model.py | 13 | except: | CODE |
| HIGH | RWKV-v4/src/model.py | 343 | except: | CODE |
| HIGH | RWKV-v5/rwkv_v6_demo.py | 653 | except: | CODE |
| HIGH | RWKV-v5/rwkv_v6_demo.py | 632 | except: | CODE |
| HIGH | RWKV-v5/tokenizer/rwkv_tokenizer.py | 100 | except: | CODE |
| HIGH | RWKV-v5/src/trainer.py | 135 | except: | CODE |
| HIGH | RWKV-v5/src/trainer.py | 109 | except: | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | RWKV-v7/rwkv_v7_demo.py | 5 | CODE | |
| LOW | RWKV-v7/rwkv_v7_demo.py | 5 | CODE | |
| LOW | RWKV-v7/run_qwen35_make_pth.py | 12 | CODE | |
| LOW | RWKV-v7/rwkv_mmlu_eval.py | 4 | CODE | |
| LOW | RWKV-v7/rwkv_mmlu_eval.py | 4 | CODE | |
| LOW | RWKV-v7/rwkv_mmlu_eval.py | 14 | CODE | |
| LOW | RWKV-v7/train_temp/rwkv7_train_simplified.py | 5 | CODE | |
| LOW | RWKV-v7/train_temp/train.py | 11 | CODE | |
| LOW | RWKV-v7/train_temp/train.py | 65 | CODE | |
| LOW | RWKV-v7/train_temp/train.py | 65 | CODE | |
| LOW | RWKV-v7/train_temp/train.py | 65 | CODE | |
| LOW | RWKV-v7/train_temp/src/model.py | 5 | CODE | |
| LOW | RWKV-v7/train_temp/src/model.py | 8 | CODE | |
| LOW | RWKV-v7/train_temp/src/model.py | 10 | CODE | |
| LOW | RWKV-v7/train_temp/src/model.py | 10 | CODE | |
| LOW | RWKV-v7/train_temp/src/dataset.py | 5 | CODE | |
| LOW | RWKV-v7/train_temp/src/dataset.py | 5 | CODE | |
| LOW | RWKV-v7/train_temp/src/dataset.py | 5 | CODE | |
| LOW | RWKV-v7/train_temp/src/dataset.py | 5 | CODE | |
| LOW | RWKV-v7/train_temp/src/dataset.py | 6 | CODE | |
| LOW | RWKV-v7/train_temp/src/trainer.py | 1 | CODE | |
| LOW | RWKV-v7/train_temp/src/trainer.py | 1 | CODE | |
| LOW | RWKV-v7/train_temp/src/trainer.py | 3 | CODE | |
| LOW | RWKV-v7/train_temp/src/trainer.py | 5 | CODE | |
| LOW | RWKV-v8/251016_rosa_1bit_run.py | 3 | CODE | |
| LOW | RWKV-v8/251018_rosa_4bit_run.py | 3 | CODE | |
| LOW | RWKV-v8/260222_rosa4bitLM_L12.py | 5 | CODE | |
| LOW | RWKV-v8/260222_rosa4bitLM_L12.py | 5 | CODE | |
| LOW | RWKV-v8/260222_rosa4bitLM_L12.py | 5 | CODE | |
| LOW | RWKV-v8/260222_rosa4bitLM_L12.py | 5 | CODE | |
| LOW | RWKV-v8/260212_rosa1bitLM_L12.py | 5 | CODE | |
| LOW | RWKV-v8/260212_rosa1bitLM_L12.py | 5 | CODE | |
| LOW | RWKV-v8/260212_rosa1bitLM_L12.py | 5 | CODE | |
| LOW | RWKV-v8/260212_rosa1bitLM_L12.py | 5 | CODE | |
| LOW | RWKV-v1/train.py | 5 | CODE | |
| LOW | RWKV-v1/train.py | 5 | CODE | |
| LOW | RWKV-v1/train.py | 5 | CODE | |
| LOW | RWKV-v1/train.py | 5 | CODE | |
| LOW | RWKV-v1/src/utils.py | 4 | CODE | |
| LOW | RWKV-v1/src/trainer.py | 3 | CODE | |
| LOW | RWKV-v1/src/trainer.py | 6 | CODE | |
| LOW | RWKV-v1/src/trainer.py | 7 | CODE | |
| LOW | RWKV-v4neo/img_demoAE.py | 5 | CODE | |
| LOW | RWKV-v4neo/run.py | 6 | CODE | |
| LOW | RWKV-v4neo/train.py | 11 | CODE | |
| LOW | RWKV-v4neo/train.py | 135 | CODE | |
| LOW | RWKV-v4neo/train.py | 135 | CODE | |
| LOW | RWKV-v4neo/train.py | 135 | CODE | |
| LOW | RWKV-v4neo/src/model_run.py | 7 | CODE | |
| LOW | RWKV-v4neo/src/model_run.py | 10 | CODE | |
| LOW | RWKV-v4neo/src/model_run.py | 10 | CODE | |
| LOW | RWKV-v4neo/src/model.py | 12 | CODE | |
| LOW | RWKV-v4neo/src/model.py | 12 | CODE | |
| LOW | RWKV-v4neo/src/dataset.py | 5 | CODE | |
| LOW | RWKV-v4neo/src/dataset.py | 5 | CODE | |
| LOW | RWKV-v4neo/src/model_img.py | 6 | CODE | |
| LOW | RWKV-v4neo/src/model_img.py | 12 | CODE | |
| LOW | RWKV-v4neo/src/model_img.py | 12 | CODE | |
| LOW | RWKV-v4neo/src/model_img.py | 14 | CODE | |
| LOW | RWKV-v4neo/src/binidx.py | 1 | CODE | |
| 65 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | RWKV-v7/rwkv_v7a_demo.py | 345 | CODE | |
| LOW | RWKV-v7/rwkv_v8_rc00_demo.py | 339 | CODE | |
| LOW | RWKV-v7/rwkv_v7_demo_fast.py | 334 | CODE | |
| LOW | RWKV-v7/rwkv_v7_demo_rnn.py | 234 | CODE | |
| LOW | RWKV-v7/rwkv_v7_demo.py | 88 | CODE | |
| LOW | RWKV-v7/rwkv_v7b_demo.py | 311 | CODE | |
| LOW | RWKV-v7/rwkv_v8_rc00_hybrid_demo.py | 295 | CODE | |
| LOW | RWKV-v7/train_temp/rwkv7_train_simplified.py | 67 | CODE | |
| LOW | RWKV-v7/train_temp/rwkv7_train_simplified.py | 89 | CODE | |
| LOW | RWKV-v7/train_temp/src/model.py | 497 | CODE | |
| LOW | RWKV-v7/train_temp/src/model.py | 892 | CODE | |
| LOW | RWKV-v7/train_temp/src/model.py | 524 | CODE | |
| LOW | RWKV-v7/train_temp/src/trainer.py | 159 | CODE | |
| LOW | RWKV-v7/train_temp/src/trainer.py | 130 | CODE | |
| LOW | RWKV-v8/251105_reverse_run.py | 41 | CODE | |
| LOW | RWKV-v8/251105_reverse_run.py | 145 | CODE | |
| LOW | RWKV-v8/251105_reverse_run.py | 167 | CODE | |
| LOW | RWKV-v8/251016_rosa_1bit_run.py | 9 | CODE | |
| LOW | RWKV-v8/251024_rosaQKV_run.py | 29 | CODE | |
| LOW | RWKV-v8/251024_rosaQKV_run.py | 133 | CODE | |
| LOW | RWKV-v8/251024_rosaQKV_run.py | 155 | CODE | |
| LOW | RWKV-v8/251014_rosa_onlyemb_train.py | 9 | CODE | |
| LOW | RWKV-v8/251014_rosa_1bit_layer.py | 6 | CODE | |
| LOW | RWKV-v8/251014_rosa_1bit_layer.py | 42 | CODE | |
| LOW | RWKV-v8/251014_rosa_1bit_train.py | 9 | CODE | |
| LOW | RWKV-v8/251014_rosa_1bit_train.py | 61 | CODE | |
| LOW | RWKV-v8/251018_rosa_4bit_run.py | 9 | CODE | |
| LOW | RWKV-v8/260222_rosa4bitLM_L12.py | 156 | CODE | |
| LOW | RWKV-v8/260222_rosa4bitLM_L12.py | 84 | CODE | |
| LOW | RWKV-v8/260222_rosa4bitLM_L12.py | 179 | CODE | |
| LOW | RWKV-v8/260212_rosa1bitLM_L12.py | 150 | CODE | |
| LOW | RWKV-v8/260212_rosa1bitLM_L12.py | 78 | CODE | |
| LOW | RWKV-v1/src/model.py | 16 | CODE | |
| LOW | RWKV-v1/src/model.py | 392 | CODE | |
| LOW | RWKV-v1/src/model.py | 461 | CODE | |
| LOW | RWKV-v1/src/trainer.py | 60 | CODE | |
| LOW | RWKV-v1/src/trainer.py | 65 | CODE | |
| LOW | RWKV-v4neo/chat.py | 197 | CODE | |
| LOW | RWKV-v4neo/math_demo/run.py | 41 | CODE | |
| LOW | RWKV-v4neo/src/model_run.py | 36 | CODE | |
| LOW | RWKV-v4neo/src/model.py | 747 | CODE | |
| LOW | RWKV-v4neo/src/model.py | 822 | CODE | |
| LOW | RWKV-v4neo/src/model.py | 913 | CODE | |
| LOW | RWKV-v4neo/src/dataset.py | 15 | CODE | |
| LOW | RWKV-v4neo/src/dataset.py | 107 | CODE | |
| LOW | RWKV-v4neo/src/model_img.py | 295 | CODE | |
| LOW | RWKV-v4neo/src/trainer.py | 203 | CODE | |
| LOW | RWKV-v4neo/src/trainer.py | 171 | CODE | |
| LOW | RWKV-v3/src/model_run.py | 182 | CODE | |
| LOW | RWKV-v3/src/model.py | 62 | CODE | |
| LOW | RWKV-v3/src/trainer.py | 78 | CODE | |
| LOW | RWKV-v3/src/trainer.py | 83 | CODE | |
| LOW | RWKV-v2-RNN/src/model_run.py | 14 | CODE | |
| LOW | RWKV-v2-RNN/src/model.py | 63 | CODE | |
| LOW | RWKV-v2-RNN/src/trainer.py | 78 | CODE | |
| LOW | RWKV-v2-RNN/src/trainer.py | 83 | CODE | |
| LOW | RWKV-v4/src/model_run.py | 211 | CODE | |
| LOW | RWKV-v4/src/model_run.py | 241 | CODE | |
| LOW | RWKV-v4/src/model.py | 107 | CODE | |
| LOW | RWKV-v4/src/model.py | 382 | CODE | |
| 15 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | RWKV-v7/train_temp/train.py | 0 | ############################################################################ # # rwkv-5 {args.precision.upper()} on {arg | STRING |
| HIGH | RWKV-v4neo/train.py | 0 | ############################################################################ # # rwkv-5 {args.precision.upper()} on {arg | STRING |
| HIGH | RWKV-v5/train.py | 0 | ############################################################################ # # rwkv-5 {args.precision.upper()} on {arg | STRING |
| HIGH | RWKV-v7/train_temp/src/model.py | 0 | ############################################################################ # # init model weight (slow for large model | STRING |
| HIGH | RWKV-v4neo/src/model.py | 0 | ############################################################################ # # init model weight (slow for large model | STRING |
| HIGH | RWKV-v4neo/src/model_img.py | 0 | ############################################################################ # # init model weight (slow for large model | STRING |
| HIGH | RWKV-v5/src/model.py | 0 | ############################################################################ # # init model weight (slow for large model | STRING |
| HIGH | RWKV-v7/train_temp/src/binidx.py | 0 | if distributed is initialized print only on rank 0. | STRING |
| HIGH | RWKV-v4neo/src/binidx.py | 0 | if distributed is initialized print only on rank 0. | STRING |
| HIGH | RWKV-v4/src/binidx.py | 0 | if distributed is initialized print only on rank 0. | STRING |
| HIGH | RWKV-v5/src/binidx.py | 0 | if distributed is initialized print only on rank 0. | STRING |
| HIGH | RWKV-v7/train_temp/src/binidx.py | 0 | retrieves a single item from the dataset with the option to only return a portion of the item. get(idx) is the same as [ | STRING |
| HIGH | RWKV-v4neo/src/binidx.py | 0 | retrieves a single item from the dataset with the option to only return a portion of the item. get(idx) is the same as [ | STRING |
| HIGH | RWKV-v4/src/binidx.py | 0 | retrieves a single item from the dataset with the option to only return a portion of the item. get(idx) is the same as [ | STRING |
| HIGH | RWKV-v5/src/binidx.py | 0 | retrieves a single item from the dataset with the option to only return a portion of the item. get(idx) is the same as [ | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | RWKV-v4neo/run.py | 20 | # Step 1: set model & config (use v4 to run your trained-from-scratch models. v4 and v4neo are compatible) | COMMENT |
| LOW | RWKV-v4neo/run.py | 78 | # Step 2: set prompt & sampling stuffs | COMMENT |
| LOW | RWKV-v3/run.py | 19 | ### Step 1: set model ################################################################################## | COMMENT |
| LOW⚡ | RWKV-v3/run.py | 37 | ### Step 2: set context ################################################################################ | COMMENT |
| LOW⚡ | RWKV-v3/train.py | 28 | ### Step 1: set training data ########################################################################## | COMMENT |
| LOW⚡ | RWKV-v3/train.py | 34 | ### Step 2: set model size ############################################################################# | COMMENT |
| LOW⚡ | RWKV-v3/train.py | 48 | ### Step 3: set batch size ############################################################################# | COMMENT |
| LOW⚡ | RWKV-v3/train.py | 55 | ### Step 4: set learning rate, number of mini-epochs ####################################################### | COMMENT |
| LOW | RWKV-v2-RNN/run.py | 20 | ### Step 1: set model ################################################################################## | COMMENT |
| LOW⚡ | RWKV-v2-RNN/run.py | 44 | ### Step 2: set context ################################################################################ | COMMENT |
| LOW⚡ | RWKV-v2-RNN/train.py | 17 | ### Step 1: set training data ########################################################################## | COMMENT |
| LOW⚡ | RWKV-v2-RNN/train.py | 23 | ### Step 2: set model size ############################################################################# | COMMENT |
| LOW⚡ | RWKV-v2-RNN/train.py | 32 | ### Step 3: set batch size ############################################################################# | COMMENT |
| LOW⚡ | RWKV-v2-RNN/train.py | 39 | ### Step 4: set learning rate, training mini-epochs ####################################################### | COMMENT |
| LOW⚡ | RWKV-v4/run.py | 19 | # Step 1: set model | COMMENT |
| LOW | RWKV-v4/run.py | 70 | # Step 2: set prompt & sampling stuffs | COMMENT |
| LOW | RWKV-v4/train.py | 22 | # Step 1: set training data & cfg | COMMENT |
| LOW | RWKV-v4/train.py | 78 | # Step 2: set model details | COMMENT |
| LOW | RWKV-v4/train.py | 110 | # Step 3: set batch size & learning rate etc. | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | RWKV-v4neo/run.py | 20 | # Step 1: set model & config (use v4 to run your trained-from-scratch models. v4 and v4neo are compatible) | COMMENT |
| LOW | RWKV-v4neo/run.py | 78 | # Step 2: set prompt & sampling stuffs | COMMENT |
| LOW | RWKV-v3/run.py | 19 | ### Step 1: set model ################################################################################## | COMMENT |
| LOW⚡ | RWKV-v3/run.py | 37 | ### Step 2: set context ################################################################################ | COMMENT |
| LOW⚡ | RWKV-v3/train.py | 28 | ### Step 1: set training data ########################################################################## | COMMENT |
| LOW⚡ | RWKV-v3/train.py | 34 | ### Step 2: set model size ############################################################################# | COMMENT |
| LOW⚡ | RWKV-v3/train.py | 48 | ### Step 3: set batch size ############################################################################# | COMMENT |
| LOW⚡ | RWKV-v3/train.py | 55 | ### Step 4: set learning rate, number of mini-epochs ####################################################### | COMMENT |
| LOW | RWKV-v2-RNN/run.py | 20 | ### Step 1: set model ################################################################################## | COMMENT |
| LOW⚡ | RWKV-v2-RNN/run.py | 44 | ### Step 2: set context ################################################################################ | COMMENT |
| LOW⚡ | RWKV-v2-RNN/train.py | 17 | ### Step 1: set training data ########################################################################## | COMMENT |
| LOW⚡ | RWKV-v2-RNN/train.py | 23 | ### Step 2: set model size ############################################################################# | COMMENT |
| LOW⚡ | RWKV-v2-RNN/train.py | 32 | ### Step 3: set batch size ############################################################################# | COMMENT |
| LOW⚡ | RWKV-v2-RNN/train.py | 39 | ### Step 4: set learning rate, training mini-epochs ####################################################### | COMMENT |
| LOW⚡ | RWKV-v4/run.py | 19 | # Step 1: set model | COMMENT |
| LOW | RWKV-v4/run.py | 70 | # Step 2: set prompt & sampling stuffs | COMMENT |
| LOW | RWKV-v4/train.py | 22 | # Step 1: set training data & cfg | COMMENT |
| LOW | RWKV-v4/train.py | 78 | # Step 2: set model details | COMMENT |
| LOW | RWKV-v4/train.py | 110 | # Step 3: set batch size & learning rate etc. | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | RWKV-v7/train_temp/demo-training-run.sh | 1 | #!/bin/bash | COMMENT |
| LOW | RWKV-v7/train_temp/demo-training-run.sh | 21 | # | COMMENT |
| LOW | RWKV-v7/train_temp/demo-training-run-v7-pile.sh | 1 | #!/bin/bash | COMMENT |
| LOW | RWKV-v7/train_temp/demo-training-prepare-v7-pile.sh | 1 | #!/bin/bash | COMMENT |
| LOW | RWKV-v7/train_temp/train.py | 141 | rank_zero_info( | COMMENT |
| LOW | RWKV-v7/train_temp/demo-training-prepare.sh | 1 | #!/bin/bash | COMMENT |
| LOW | RWKV-v7/train_temp/src/model.py | 581 | self.output.weight.data.zero_() | COMMENT |
| LOW | RWKV-v7/train_temp/src/model.py | 621 | # much faster CUDA version | COMMENT |
| LOW | RWKV-v7/train_temp/src/model.py | 641 | # x = RWKV7_CLAMPW_CUDA(r, w, k, v, -kk, kk*a) | COMMENT |
| LOW | RWKV-v7/train_temp/src/model.py | 681 | # self.layer_id = layer_id | COMMENT |
| LOW | RWKV-v4neo/run.py | 41 | # MODEL_NAME = "/fsx/BlinkDL/rwkv-release/RWKV-4-Pile-169M-20220807-8023" | COMMENT |
| LOW | RWKV-v4neo/run.py | 81 | # context = 'A' | COMMENT |
| LOW | RWKV-v4neo/train.py | 21 | # --data_file "" --data_type "dummy" --vocab_size 0 \ | COMMENT |
| LOW | RWKV-v4neo/train.py | 41 | # --micro_bsz 8 --n_layer 24 --n_embd 2048 --pre_ffn 0 --head_qk 0 \ | COMMENT |
| LOW | RWKV-v4neo/train.py | 261 | # | COMMENT |
| LOW | RWKV-v4neo/src/model.py | 881 | # if sum_mask == 0: | COMMENT |
| LOW | RWKV-v4neo/src/model_img.py | 421 | ss = n.split('.') | COMMENT |
| LOW | RWKV-v3/train.py | 41 | # 'RWKV' (better for English) or 'RWKV-ffnPre' (better in some cases) | COMMENT |
| LOW | RWKV-v4/train.py | 41 | COMMENT | |
| LOW | RWKV-v5/demo-training-run.sh | 1 | #!/bin/bash | COMMENT |
| LOW | RWKV-v5/demo-training-run.sh | 21 | # | COMMENT |
| LOW | RWKV-v5/rwkv_v6_demo.py | 221 | COMMENT | |
| LOW | RWKV-v5/rwkv_v6_demo.py | 241 | # att.output.weight => zero | COMMENT |
| LOW | RWKV-v5/demo-training-run-v7-pile.sh | 1 | #!/bin/bash | COMMENT |
| LOW | RWKV-v5/demo-training-prepare-v7-pile.sh | 1 | #!/bin/bash | COMMENT |
| LOW | RWKV-v5/train.py | 321 | data_loader = DataLoader(train_data, shuffle=False, pin_memory=True, batch_size=args.micro_bsz, num_workers=1, persi | COMMENT |
| LOW | RWKV-v5/demo-training-prepare.sh | 1 | #!/bin/bash | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | RWKV-v1/src/model.py | 10 | logger = logging.getLogger(__name__) | CODE |
| LOW | RWKV-v1/src/trainer.py | 9 | logger = logging.getLogger(__name__) | CODE |
| LOW | RWKV-v3/src/model.py | 12 | logger = logging.getLogger(__name__) | CODE |
| LOW | RWKV-v3/src/trainer.py | 22 | logger = logging.getLogger(__name__) | CODE |
| LOW | RWKV-v2-RNN/src/model.py | 12 | logger = logging.getLogger(__name__) | CODE |
| LOW | RWKV-v2-RNN/src/trainer.py | 22 | logger = logging.getLogger(__name__) | CODE |
| LOW | RWKV-v4/src/model.py | 16 | logger = logging.getLogger(__name__) | CODE |
| LOW | RWKV-v4/src/trainer.py | 18 | logger = logging.getLogger(__name__) | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | RWKV-v7/train_temp/demo-training-run.sh | 8 | # Therefore check the log (### Loading rwkv-xxx.pth... ###), and make sure you don't have extra rwkv-*.pth there | COMMENT |
| LOW | RWKV-v7/train_temp/demo-training-run-v7-pile.sh | 8 | # Therefore check the log (### Loading rwkv-xxx.pth... ###), and make sure you don't have extra rwkv-*.pth there | COMMENT |
| LOW | RWKV-v5/demo-training-run.sh | 8 | # Therefore check the log (### Loading rwkv-xxx.pth... ###), and make sure you don't have extra rwkv-*.pth there | COMMENT |
| LOW | RWKV-v5/demo-training-run-v7-pile.sh | 8 | # Therefore check the log (### Loading rwkv-xxx.pth... ###), and make sure you don't have extra rwkv-*.pth there | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | RWKV-v3/run.py | 32 | UNKNOWN_CHAR = ' ' # here we just set it to [space] for simplicity | CODE |
| LOW⚡ | RWKV-v2-RNN/run.py | 39 | UNKNOWN_CHAR = ' ' # here we just set it to [space] for simplicity | CODE |
| LOW | RWKV-v4/run.py | 36 | UNKNOWN_CHAR = ' ' # here we just set it to ' ' for simplicity | CODE |
| LOW | RWKV-v4/verify.py | 28 | UNKNOWN_CHAR = ' ' # here we just set it to [space] for simplicity | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | RWKV-v4/run.py | 21 | # Set TOKEN_MODE to 'char' or 'bpe' if the model is trained by 'train.py' from scratch. | COMMENT |
| LOW⚡ | RWKV-v4/run.py | 23 | # Set TOKEN_MODE to 'pile' if you want to test pre-trained pile models. | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | RWKV-v8/251105_reverse_run.py | 83 | CODE | |
| LOW | RWKV-v8/251024_rosaQKV_run.py | 71 | CODE | |
| LOW | RWKV-v8/260212_rosa1bitLM_L12.py | 192 | CODE | |
| LOW | RWKV-v4neo/src/model_img.py | 370 | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | RWKV-v7/train_temp/src/trainer.py | 148 | except Exception as e: | CODE |
| LOW | RWKV-v4neo/src/trainer.py | 189 | except Exception as e: | CODE |
| LOW | RWKV-v5/src/trainer.py | 196 | except Exception as e: | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | RWKV-v7/run_qwen35_make_pth.py | 179 | def strip_language_model_prefix(key: str) -> str: | CODE |