Release CoTranslator paired data and translator checkpoint
Browse files- .gitattributes +2 -0
- README.md +133 -0
- args.json +432 -0
- assets/main.png +3 -0
- chat_template.jinja +86 -0
- config.json +71 -0
- latest_checkpointed_iteration.txt +1 -0
- model-00001-of-00002.safetensors +3 -0
- model-00002-of-00002.safetensors +3 -0
- model.safetensors.index.json +405 -0
- tokenizer.json +3 -0
- tokenizer_config.json +30 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
assets/main.png filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,133 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: Qwen/Qwen3-4B-Thinking-2507
|
| 5 |
+
base_model_relation: finetune
|
| 6 |
+
pipeline_tag: text-generation
|
| 7 |
+
language:
|
| 8 |
+
- en
|
| 9 |
+
datasets:
|
| 10 |
+
- Word2Li/CoTranslator-Nemotron-SFT-Math-v4-Pairs
|
| 11 |
+
tags:
|
| 12 |
+
- cotranslator
|
| 13 |
+
- reasoning-distillation
|
| 14 |
+
- chain-of-thought
|
| 15 |
+
- math
|
| 16 |
+
- qwen3
|
| 17 |
+
- supervised-fine-tuning
|
| 18 |
+
---
|
| 19 |
+
|
| 20 |
+
# CoTranslator-4B
|
| 21 |
+
|
| 22 |
+
Main translator trained on the 1/8 paired-data subset, introduced in **CoTranslator: Reverse to Forward Chain-of-Thought Translator for Reasoning Distillation**.
|
| 23 |
+
CoTranslator learns to convert answer-conditioned reverse chain-of-thought (CoT) trajectories into trajectories resembling verified forward solutions. Its translations provide supervision for downstream student models.
|
| 24 |
+
|
| 25 |
+
|
| 26 |
+
|
| 27 |
+
*Overview of CoTranslator. The translator learns from paired reverse and verified forward trajectories, then translates reverse trajectories at scale to construct student training data.*
|
| 28 |
+
|
| 29 |
+
## Released checkpoint
|
| 30 |
+
|
| 31 |
+
| Item | Value |
|
| 32 |
+
| --- | --- |
|
| 33 |
+
| Base model | [Qwen/Qwen3-4B-Thinking-2507](https://huggingface.co/Qwen/Qwen3-4B-Thinking-2507) |
|
| 34 |
+
| Parameters stored | 4,022,468,096 |
|
| 35 |
+
| Training | Full-parameter supervised fine-tuning |
|
| 36 |
+
| Source pair configuration | [`subset_1_8`](https://huggingface.co/datasets/Word2Li/CoTranslator-Nemotron-SFT-Math-v4-Pairs) |
|
| 37 |
+
| Rows in the source JSONL | 33,286, before training preprocessing and packing |
|
| 38 |
+
| Checkpoint step | 616 |
|
| 39 |
+
| Weights | BF16, two Safetensors shards |
|
| 40 |
+
|
| 41 |
+
The companion translator is [CoTranslator-4B-Full](https://huggingface.co/Word2Li/CoTranslator-4B-Full).
|
| 42 |
+
|
| 43 |
+
## Translation task
|
| 44 |
+
|
| 45 |
+
The teacher generates a reverse trajectory with access to the reference answer, using the semantic suppression prompt described in the manuscript. The translator receives that trajectory, including its reasoning and final response. During training, its target is the corresponding question-only forward response from the original Nemotron dataset, accepted by answer verification.
|
| 46 |
+
|
| 47 |
+
The translation input consists of the following system instruction and a user message containing the reverse trajectory:
|
| 48 |
+
|
| 49 |
+
```text
|
| 50 |
+
You are a reasoning trajectory refiner. Given a teacher-generated solution trajectory, produce a more natural reasoning trajectory. Preserve the mathematical conclusion and answer format.
|
| 51 |
+
```
|
| 52 |
+
|
| 53 |
+
```text
|
| 54 |
+
<think>
|
| 55 |
+
{{reasoning_content}}
|
| 56 |
+
</think>
|
| 57 |
+
{{content}}
|
| 58 |
+
```
|
| 59 |
+
|
| 60 |
+
At inference, supply the teacher trajectory in this format. The manuscript conditions the translator on the trajectory alone; the problem and reference answer are used for pairing and verification.
|
| 61 |
+
|
| 62 |
+
## Training configuration
|
| 63 |
+
|
| 64 |
+
The settings below are recorded in this checkpoint's `args.json` and match the manuscript's translator configuration.
|
| 65 |
+
|
| 66 |
+
| Setting | Value |
|
| 67 |
+
| --- | --- |
|
| 68 |
+
| Framework | Megatron-Swift |
|
| 69 |
+
| Epochs | 3 |
|
| 70 |
+
| Learning rate | 5e-5 |
|
| 71 |
+
| LR schedule / warmup fraction | Cosine / 0.1 |
|
| 72 |
+
| Weight decay | 0.1 |
|
| 73 |
+
| Global / micro batch size | 32 / 2 |
|
| 74 |
+
| Sequence length | 65,536 tokens |
|
| 75 |
+
| Packing / precision | Enabled / BF16 |
|
| 76 |
+
| Adam beta1 / beta2 | 0.9 / 0.95 |
|
| 77 |
+
| Gradient clipping | 1.0 |
|
| 78 |
+
| Seed | 42 |
|
| 79 |
+
| Chat template | `qwen3_thinking` |
|
| 80 |
+
|
| 81 |
+
Training preprocessing drops examples exceeding the 65,536-token limit; the source JSONL row count precedes that preprocessing.
|
| 82 |
+
|
| 83 |
+
## Usage
|
| 84 |
+
|
| 85 |
+
Use a recent Transformers release. The checkpoint configuration was saved with Transformers 5.12.1; local configuration and tokenizer loading were checked with that version.
|
| 86 |
+
|
| 87 |
+
```bash
|
| 88 |
+
pip install -U transformers accelerate torch datasets
|
| 89 |
+
```
|
| 90 |
+
|
| 91 |
+
This example translates the input trajectory from one released training pair:
|
| 92 |
+
|
| 93 |
+
```python
|
| 94 |
+
import torch
|
| 95 |
+
from datasets import load_dataset
|
| 96 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 97 |
+
|
| 98 |
+
model_id = "Word2Li/CoTranslator-4B"
|
| 99 |
+
data = load_dataset(
|
| 100 |
+
"Word2Li/CoTranslator-Nemotron-SFT-Math-v4-Pairs", "subset_1_8",
|
| 101 |
+
split="train", streaming=True,
|
| 102 |
+
)
|
| 103 |
+
example = next(iter(data))
|
| 104 |
+
messages = example["messages"][:2]
|
| 105 |
+
|
| 106 |
+
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
| 107 |
+
model = AutoModelForCausalLM.from_pretrained(
|
| 108 |
+
model_id, dtype="auto", device_map="auto",
|
| 109 |
+
)
|
| 110 |
+
prompt = tokenizer.apply_chat_template(
|
| 111 |
+
messages, tokenize=False, add_generation_prompt=True,
|
| 112 |
+
)
|
| 113 |
+
inputs = tokenizer(prompt, add_special_tokens=False, return_tensors="pt").to(model.device)
|
| 114 |
+
with torch.inference_mode():
|
| 115 |
+
output = model.generate(
|
| 116 |
+
**inputs, max_new_tokens=32768, do_sample=True,
|
| 117 |
+
temperature=0.6, top_p=0.95, top_k=20,
|
| 118 |
+
pad_token_id=tokenizer.pad_token_id,
|
| 119 |
+
eos_token_id=tokenizer.eos_token_id,
|
| 120 |
+
)
|
| 121 |
+
continuation = tokenizer.decode(
|
| 122 |
+
output[0, inputs["input_ids"].shape[1]:], skip_special_tokens=True,
|
| 123 |
+
)
|
| 124 |
+
# The bundled chat template already places the opening <think> in the prompt.
|
| 125 |
+
translated_trajectory = "<think>\n" + continuation
|
| 126 |
+
print(translated_trajectory)
|
| 127 |
+
```
|
| 128 |
+
|
| 129 |
+
The manuscript uses temperature 0.6, top-p 0.95, top-k 20, min-p 0, and a maximum of 32,768 generated tokens for translator inference.
|
| 130 |
+
|
| 131 |
+
## License and attribution
|
| 132 |
+
|
| 133 |
+
This release uses Apache-2.0, matching the [base model license](https://huggingface.co/Qwen/Qwen3-4B-Thinking-2507/blob/main/LICENSE). Training data attribution and source-specific data licenses are described in the [paired dataset card](https://huggingface.co/datasets/Word2Li/CoTranslator-Nemotron-SFT-Math-v4-Pairs).
|
args.json
ADDED
|
@@ -0,0 +1,432 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"use_ray": false,
|
| 3 |
+
"ray_exp_name": null,
|
| 4 |
+
"device_groups": null,
|
| 5 |
+
"model": "/mnt/exp/moe-lite/tangzinan.tzn/models/Qwen3-4B-Thinking-2507",
|
| 6 |
+
"model_type": "qwen3",
|
| 7 |
+
"model_revision": null,
|
| 8 |
+
"task_type": "causal_lm",
|
| 9 |
+
"torch_dtype": "bfloat16",
|
| 10 |
+
"attn_impl": null,
|
| 11 |
+
"experts_impl": null,
|
| 12 |
+
"new_special_tokens": [],
|
| 13 |
+
"num_labels": null,
|
| 14 |
+
"problem_type": null,
|
| 15 |
+
"rope_scaling": null,
|
| 16 |
+
"device_map": null,
|
| 17 |
+
"max_memory": {},
|
| 18 |
+
"max_model_len": null,
|
| 19 |
+
"local_repo_path": null,
|
| 20 |
+
"init_strategy": null,
|
| 21 |
+
"template": "qwen3_thinking",
|
| 22 |
+
"system": null,
|
| 23 |
+
"max_length": 65536,
|
| 24 |
+
"truncation_strategy": "delete",
|
| 25 |
+
"max_pixels": null,
|
| 26 |
+
"agent_template": "hermes",
|
| 27 |
+
"norm_bbox": null,
|
| 28 |
+
"use_chat_template": true,
|
| 29 |
+
"padding_side": "right",
|
| 30 |
+
"padding_free": true,
|
| 31 |
+
"loss_scale": "default",
|
| 32 |
+
"sequence_parallel_size": 4,
|
| 33 |
+
"is_binary_loss_scale": null,
|
| 34 |
+
"template_backend": "swift",
|
| 35 |
+
"response_prefix": null,
|
| 36 |
+
"enable_thinking": null,
|
| 37 |
+
"preserve_thinking": null,
|
| 38 |
+
"add_non_thinking_prefix": true,
|
| 39 |
+
"disable_ignore_empty_think": false,
|
| 40 |
+
"dataset": [],
|
| 41 |
+
"val_dataset": [],
|
| 42 |
+
"cached_dataset": [
|
| 43 |
+
"/mnt/exp/moe-lite/tangzinan.tzn/iclr_2027/data/cache/Qwen3-4B-Thinking-2507-Nemotron-SFT-Math-v4-cot-pair-refiner-1_8-64k/train"
|
| 44 |
+
],
|
| 45 |
+
"cached_val_dataset": [],
|
| 46 |
+
"split_dataset_ratio": 0.0,
|
| 47 |
+
"data_seed": 42,
|
| 48 |
+
"dataset_num_proc": 16,
|
| 49 |
+
"load_from_cache_file": false,
|
| 50 |
+
"dataset_shuffle": true,
|
| 51 |
+
"val_dataset_shuffle": false,
|
| 52 |
+
"streaming": false,
|
| 53 |
+
"interleave_prob": null,
|
| 54 |
+
"stopping_strategy": "first_exhausted",
|
| 55 |
+
"shuffle_buffer_size": 1000,
|
| 56 |
+
"download_mode": "reuse_dataset_if_exists",
|
| 57 |
+
"columns": {},
|
| 58 |
+
"strict": false,
|
| 59 |
+
"remove_unused_columns": true,
|
| 60 |
+
"disable_auto_column_mapping": false,
|
| 61 |
+
"model_name": null,
|
| 62 |
+
"model_author": null,
|
| 63 |
+
"custom_dataset_info": [],
|
| 64 |
+
"quant_method": null,
|
| 65 |
+
"quant_bits": null,
|
| 66 |
+
"hqq_axis": null,
|
| 67 |
+
"bnb_4bit_compute_dtype": "bfloat16",
|
| 68 |
+
"bnb_4bit_quant_type": "nf4",
|
| 69 |
+
"bnb_4bit_use_double_quant": true,
|
| 70 |
+
"bnb_4bit_quant_storage": null,
|
| 71 |
+
"max_new_tokens": null,
|
| 72 |
+
"temperature": 0.9,
|
| 73 |
+
"top_k": -1,
|
| 74 |
+
"top_p": 1.0,
|
| 75 |
+
"repetition_penalty": 1.0,
|
| 76 |
+
"num_beams": 1,
|
| 77 |
+
"stream": false,
|
| 78 |
+
"stop_words": [],
|
| 79 |
+
"logprobs": false,
|
| 80 |
+
"top_logprobs": null,
|
| 81 |
+
"structured_outputs_regex": null,
|
| 82 |
+
"tuner_backend": "peft",
|
| 83 |
+
"tuner_type": "full",
|
| 84 |
+
"adapters": [],
|
| 85 |
+
"external_plugins": [],
|
| 86 |
+
"custom_register_path": [],
|
| 87 |
+
"seed": 42,
|
| 88 |
+
"model_kwargs": {},
|
| 89 |
+
"enable_npu_model_patch": true,
|
| 90 |
+
"load_args": false,
|
| 91 |
+
"load_data_args": false,
|
| 92 |
+
"packing": true,
|
| 93 |
+
"packing_length": 65536,
|
| 94 |
+
"packing_num_proc": 16,
|
| 95 |
+
"lazy_tokenize": false,
|
| 96 |
+
"use_hf": false,
|
| 97 |
+
"hub_token": null,
|
| 98 |
+
"ddp_timeout": 18000000,
|
| 99 |
+
"ddp_backend": "nccl",
|
| 100 |
+
"ignore_args_error": false,
|
| 101 |
+
"use_swift_lora": false,
|
| 102 |
+
"freeze_llm": false,
|
| 103 |
+
"freeze_vit": true,
|
| 104 |
+
"freeze_aligner": true,
|
| 105 |
+
"freeze_parameters": [],
|
| 106 |
+
"freeze_parameters_regex": null,
|
| 107 |
+
"freeze_parameters_ratio": 0.0,
|
| 108 |
+
"trainable_parameters": [],
|
| 109 |
+
"trainable_parameters_regex": null,
|
| 110 |
+
"target_modules": [
|
| 111 |
+
"all-linear"
|
| 112 |
+
],
|
| 113 |
+
"target_regex": null,
|
| 114 |
+
"modules_to_save": [],
|
| 115 |
+
"lora_rank": 8,
|
| 116 |
+
"lora_alpha": 32,
|
| 117 |
+
"lora_dropout": 0.05,
|
| 118 |
+
"lora_bias": "none",
|
| 119 |
+
"lora_dtype": null,
|
| 120 |
+
"use_rslora": false,
|
| 121 |
+
"rlhf_type": null,
|
| 122 |
+
"loss_type": null,
|
| 123 |
+
"mcore_ref_model": null,
|
| 124 |
+
"mcore_ref_adapter": null,
|
| 125 |
+
"beta": null,
|
| 126 |
+
"rpo_alpha": null,
|
| 127 |
+
"reference_free": false,
|
| 128 |
+
"label_smoothing": 0.0,
|
| 129 |
+
"f_divergence_type": "reverse_kl",
|
| 130 |
+
"desirable_weight": 1.0,
|
| 131 |
+
"undesirable_weight": 1.0,
|
| 132 |
+
"calculate_KL": null,
|
| 133 |
+
"center_rewards_coefficient": null,
|
| 134 |
+
"teacher_model": null,
|
| 135 |
+
"teacher_model_type": null,
|
| 136 |
+
"teacher_model_revision": null,
|
| 137 |
+
"_teacher_use_disable_adapter": false,
|
| 138 |
+
"teacher_model_server": null,
|
| 139 |
+
"teacher_tag_key": "dataset",
|
| 140 |
+
"gkd_logits_topk": null,
|
| 141 |
+
"lmbda": 0.5,
|
| 142 |
+
"seq_kd": false,
|
| 143 |
+
"offload_teacher_model": false,
|
| 144 |
+
"sft_alpha": 0.0,
|
| 145 |
+
"teacher_kl_coef": 1.0,
|
| 146 |
+
"generation_batch_size": null,
|
| 147 |
+
"steps_per_generation": null,
|
| 148 |
+
"num_generations": 8,
|
| 149 |
+
"num_generations_eval": null,
|
| 150 |
+
"max_completion_length": 512,
|
| 151 |
+
"importance_sampling_level": "token",
|
| 152 |
+
"tau_pos": 1.0,
|
| 153 |
+
"tau_neg": 1.05,
|
| 154 |
+
"real_tau": 0.5,
|
| 155 |
+
"fipo_decay_rate": 32.0,
|
| 156 |
+
"fipo_clip_range": 0.2,
|
| 157 |
+
"fipo_clip_high_only": true,
|
| 158 |
+
"fipo_safety_threshold": 4.0,
|
| 159 |
+
"epsilon": 0.2,
|
| 160 |
+
"epsilon_high": null,
|
| 161 |
+
"delta": null,
|
| 162 |
+
"use_vllm": true,
|
| 163 |
+
"vllm_mode": null,
|
| 164 |
+
"vllm_enable_prefix_caching": true,
|
| 165 |
+
"vllm_gpu_memory_utilization": 0.9,
|
| 166 |
+
"vllm_tensor_parallel_size": 1,
|
| 167 |
+
"vllm_max_model_len": null,
|
| 168 |
+
"vllm_enforce_eager": false,
|
| 169 |
+
"vllm_limit_mm_per_prompt": null,
|
| 170 |
+
"vllm_disable_cascade_attn": false,
|
| 171 |
+
"vllm_max_num_seqs": null,
|
| 172 |
+
"vllm_mm_processor_cache_gb": null,
|
| 173 |
+
"vllm_engine_kwargs": null,
|
| 174 |
+
"vllm_enable_lora": false,
|
| 175 |
+
"sleep_level": 0,
|
| 176 |
+
"offload_optimizer": false,
|
| 177 |
+
"offload_model": false,
|
| 178 |
+
"offload_bridge": false,
|
| 179 |
+
"vllm_server_base_url": null,
|
| 180 |
+
"vllm_server_host": null,
|
| 181 |
+
"vllm_server_port": [
|
| 182 |
+
8000
|
| 183 |
+
],
|
| 184 |
+
"vllm_server_timeout": 240.0,
|
| 185 |
+
"vllm_server_group_port": null,
|
| 186 |
+
"reward_funcs": [],
|
| 187 |
+
"reward_weights": null,
|
| 188 |
+
"cosine_min_len_value_wrong": -0.5,
|
| 189 |
+
"cosine_max_len_value_wrong": 0.0,
|
| 190 |
+
"cosine_min_len_value_correct": 1.0,
|
| 191 |
+
"cosine_max_len_value_correct": 0.5,
|
| 192 |
+
"cosine_max_len": null,
|
| 193 |
+
"repetition_n_grams": 3,
|
| 194 |
+
"repetition_max_penalty": -1.0,
|
| 195 |
+
"soft_max_length": null,
|
| 196 |
+
"soft_cache_length": null,
|
| 197 |
+
"dynamic_sample": false,
|
| 198 |
+
"max_resample_times": 3,
|
| 199 |
+
"overlong_filter": false,
|
| 200 |
+
"scale_rewards": "group",
|
| 201 |
+
"advantage_estimator": "grpo",
|
| 202 |
+
"kl_in_reward": false,
|
| 203 |
+
"wandb_log_unique_prompts": null,
|
| 204 |
+
"log_completions": false,
|
| 205 |
+
"rollout_importance_sampling_mode": null,
|
| 206 |
+
"rollout_importance_sampling_threshold": 2.0,
|
| 207 |
+
"log_rollout_offpolicy_metrics": false,
|
| 208 |
+
"off_policy_sequence_mask_delta": null,
|
| 209 |
+
"log_entropy": false,
|
| 210 |
+
"top_entropy_quantile": 1.0,
|
| 211 |
+
"router_replay_mode": "disabled",
|
| 212 |
+
"reward_model": null,
|
| 213 |
+
"reward_model_plugin": null,
|
| 214 |
+
"sync_ref_model": false,
|
| 215 |
+
"ref_model_sync_steps": 512,
|
| 216 |
+
"ref_model_mixup_alpha": 0.6,
|
| 217 |
+
"async_generate": false,
|
| 218 |
+
"move_model_batches": null,
|
| 219 |
+
"multi_turn_scheduler": null,
|
| 220 |
+
"max_turns": null,
|
| 221 |
+
"completion_length_limit_scope": "per_round",
|
| 222 |
+
"vllm_server_pass_dataset": false,
|
| 223 |
+
"use_gym_env": null,
|
| 224 |
+
"gym_env": null,
|
| 225 |
+
"num_iterations": 1,
|
| 226 |
+
"micro_batch_size": 2,
|
| 227 |
+
"global_batch_size": 32,
|
| 228 |
+
"recompute_granularity": "full",
|
| 229 |
+
"recompute_method": "block",
|
| 230 |
+
"recompute_num_layers": 29,
|
| 231 |
+
"recompute_modules": [
|
| 232 |
+
"core_attn"
|
| 233 |
+
],
|
| 234 |
+
"train_iters": null,
|
| 235 |
+
"num_train_epochs": 3,
|
| 236 |
+
"masked_softmax_fusion": true,
|
| 237 |
+
"bias_dropout_fusion": true,
|
| 238 |
+
"bias_activation_fusion": true,
|
| 239 |
+
"apply_rope_fusion": false,
|
| 240 |
+
"gradient_accumulation_fusion": true,
|
| 241 |
+
"cross_entropy_loss_fusion": true,
|
| 242 |
+
"cross_entropy_fusion_impl": "te",
|
| 243 |
+
"calculate_per_token_loss": true,
|
| 244 |
+
"attention_backend": "<AttnBackend.flash: 1>",
|
| 245 |
+
"optimizer": "adam",
|
| 246 |
+
"optimizer_cpu_offload": false,
|
| 247 |
+
"optimizer_offload_fraction": 1.0,
|
| 248 |
+
"optimizer_cuda_graph": false,
|
| 249 |
+
"use_precision_aware_optimizer": false,
|
| 250 |
+
"main_grads_dtype": "float32",
|
| 251 |
+
"main_params_dtype": "float32",
|
| 252 |
+
"exp_avg_dtype": "float32",
|
| 253 |
+
"exp_avg_sq_dtype": "float32",
|
| 254 |
+
"manual_gc": true,
|
| 255 |
+
"manual_gc_steps": 50,
|
| 256 |
+
"manual_gc_eval": true,
|
| 257 |
+
"train_dataloader_shuffle": true,
|
| 258 |
+
"dataloader_num_workers": 8,
|
| 259 |
+
"dataloader_pin_memory": true,
|
| 260 |
+
"dataloader_persistent_workers": true,
|
| 261 |
+
"dataloader_prefetch_factor": 2,
|
| 262 |
+
"data_sharding": false,
|
| 263 |
+
"group_by_length": false,
|
| 264 |
+
"te_rng_tracker": false,
|
| 265 |
+
"data_parallel_random_init": false,
|
| 266 |
+
"mlp_padding_free": false,
|
| 267 |
+
"lr_warmup_init": 0.0,
|
| 268 |
+
"lr": 5e-05,
|
| 269 |
+
"lr_decay_style": "cosine",
|
| 270 |
+
"lr_decay_iters": null,
|
| 271 |
+
"lr_warmup_iters": 0,
|
| 272 |
+
"lr_warmup_fraction": 0.1,
|
| 273 |
+
"min_lr": 0,
|
| 274 |
+
"lr_wsd_decay_style": "exponential",
|
| 275 |
+
"lr_wsd_decay_iters": null,
|
| 276 |
+
"weight_decay": 0.1,
|
| 277 |
+
"weight_decay_incr_style": "constant",
|
| 278 |
+
"start_weight_decay": 0.1,
|
| 279 |
+
"end_weight_decay": 0.1,
|
| 280 |
+
"clip_grad": 1.0,
|
| 281 |
+
"adam_beta1": 0.9,
|
| 282 |
+
"adam_beta2": 0.95,
|
| 283 |
+
"adam_eps": 1e-08,
|
| 284 |
+
"sgd_momentum": 0.9,
|
| 285 |
+
"muon_momentum": 0.9,
|
| 286 |
+
"muon_split_qkv": true,
|
| 287 |
+
"muon_use_nesterov": false,
|
| 288 |
+
"muon_scale_mode": "spectral",
|
| 289 |
+
"muon_fp32_matmul_prec": "medium",
|
| 290 |
+
"muon_coefficient_type": "quintic",
|
| 291 |
+
"muon_num_ns_steps": 5,
|
| 292 |
+
"muon_tp_mode": "blockwise",
|
| 293 |
+
"muon_extra_scale_factor": 1.0,
|
| 294 |
+
"muon_scalar_optimizer": "adam",
|
| 295 |
+
"output_dir": "/mnt/exp/moe-lite/tangzinan.tzn/iclr_2027/train/saves/Qwen3-4B-Thinking-2507/Nemotron-SFT-Math-v4-cot-pair-refiner-1_8-SFT",
|
| 296 |
+
"save_steps": 500,
|
| 297 |
+
"no_save_optim": false,
|
| 298 |
+
"no_save_rng": false,
|
| 299 |
+
"mcore_model": null,
|
| 300 |
+
"mcore_adapter": null,
|
| 301 |
+
"no_load_optim": false,
|
| 302 |
+
"no_load_rng": false,
|
| 303 |
+
"finetune": true,
|
| 304 |
+
"perform_initialization": false,
|
| 305 |
+
"use_cpu_initialization": false,
|
| 306 |
+
"async_save": true,
|
| 307 |
+
"save_total_limit": null,
|
| 308 |
+
"metric_for_best_model": "loss",
|
| 309 |
+
"greater_is_better": false,
|
| 310 |
+
"use_persistent_ckpt_worker": false,
|
| 311 |
+
"dist_ckpt_save_pre_mcore_014": false,
|
| 312 |
+
"dist_ckpt_optim_fully_reshardable": false,
|
| 313 |
+
"distrib_optim_fully_reshardable_mem_efficient": false,
|
| 314 |
+
"local_rank": 0,
|
| 315 |
+
"use_distributed_optimizer": true,
|
| 316 |
+
"tensor_model_parallel_size": 1,
|
| 317 |
+
"pipeline_model_parallel_size": 1,
|
| 318 |
+
"decoder_first_pipeline_num_layers": null,
|
| 319 |
+
"decoder_last_pipeline_num_layers": null,
|
| 320 |
+
"account_for_embedding_in_pipeline_split": false,
|
| 321 |
+
"account_for_loss_in_pipeline_split": false,
|
| 322 |
+
"overlap_p2p_comm": false,
|
| 323 |
+
"batch_p2p_comm": true,
|
| 324 |
+
"align_param_gather": false,
|
| 325 |
+
"sequence_parallel": false,
|
| 326 |
+
"context_parallel_size": 4,
|
| 327 |
+
"tp_comm_overlap": false,
|
| 328 |
+
"overlap_grad_reduce": true,
|
| 329 |
+
"overlap_param_gather": true,
|
| 330 |
+
"overlap_param_gather_with_optimizer_step": false,
|
| 331 |
+
"align_grad_reduce": true,
|
| 332 |
+
"virtual_pipeline_model_parallel_size": null,
|
| 333 |
+
"microbatch_group_size_per_vp_stage": null,
|
| 334 |
+
"pipeline_model_parallel_layout": null,
|
| 335 |
+
"expert_model_parallel_size": 1,
|
| 336 |
+
"expert_tensor_parallel_size": 1,
|
| 337 |
+
"report_to": [
|
| 338 |
+
"swanlab"
|
| 339 |
+
],
|
| 340 |
+
"logging_steps": 5,
|
| 341 |
+
"tensorboard_dir": "/mnt/exp/moe-lite/tangzinan.tzn/iclr_2027/train/saves/Qwen3-4B-Thinking-2507/Nemotron-SFT-Math-v4-cot-pair-refiner-1_8-SFT/runs",
|
| 342 |
+
"tensorboard_queue_size": 50,
|
| 343 |
+
"wandb_project": "megatron-swift",
|
| 344 |
+
"wandb_exp_name": null,
|
| 345 |
+
"swanlab_project": "iclr_2027",
|
| 346 |
+
"swanlab_exp_name": "Qwen3-4B-Thinking-2507-Nemotron-SFT-Math-v4-cot-pair-refiner-1_8-SFT",
|
| 347 |
+
"eval_iters": -1,
|
| 348 |
+
"eval_steps": 500,
|
| 349 |
+
"fp8_format": null,
|
| 350 |
+
"fp8_recipe": "delayed",
|
| 351 |
+
"fp8_param_gather": false,
|
| 352 |
+
"fp8_amax_history_len": 1024,
|
| 353 |
+
"fp8_amax_compute_algo": "max",
|
| 354 |
+
"fp4_format": null,
|
| 355 |
+
"fp4_recipe": "nvfp4",
|
| 356 |
+
"fp4_param_gather": false,
|
| 357 |
+
"fp16": false,
|
| 358 |
+
"bf16": true,
|
| 359 |
+
"apply_query_key_layer_scaling": false,
|
| 360 |
+
"attention_softmax_in_fp32": true,
|
| 361 |
+
"accumulate_allreduce_grads_in_fp32": true,
|
| 362 |
+
"moe_router_load_balancing_type": null,
|
| 363 |
+
"moe_router_dtype": "fp32",
|
| 364 |
+
"moe_token_dispatcher_type": null,
|
| 365 |
+
"moe_enable_deepep": false,
|
| 366 |
+
"moe_grouped_gemm": true,
|
| 367 |
+
"moe_permute_fusion": false,
|
| 368 |
+
"moe_aux_loss_coeff": null,
|
| 369 |
+
"moe_z_loss_coeff": null,
|
| 370 |
+
"moe_shared_expert_overlap": false,
|
| 371 |
+
"moe_layer_recompute": false,
|
| 372 |
+
"moe_expert_capacity_factor": null,
|
| 373 |
+
"moe_pad_expert_input_to_capacity": false,
|
| 374 |
+
"moe_token_drop_policy": null,
|
| 375 |
+
"mtp_num_layers": null,
|
| 376 |
+
"mtp_loss_scaling_factor": 0.1,
|
| 377 |
+
"mtp_decoder_input_detach": false,
|
| 378 |
+
"mtp_shared_weights": false,
|
| 379 |
+
"save_safetensors": true,
|
| 380 |
+
"ref_model": null,
|
| 381 |
+
"ref_adapters": [],
|
| 382 |
+
"merge_lora": true,
|
| 383 |
+
"max_shard_size": "5GB",
|
| 384 |
+
"vit_gradient_checkpointing": false,
|
| 385 |
+
"vit_gradient_checkpointing_kwargs": null,
|
| 386 |
+
"vit_attn_impl": null,
|
| 387 |
+
"vit_lr": null,
|
| 388 |
+
"aligner_lr": null,
|
| 389 |
+
"dsa_indexer_loss_coeff": 0.0,
|
| 390 |
+
"dsa_indexer_use_sparse_loss": false,
|
| 391 |
+
"apply_dsa_kernel_fusion": false,
|
| 392 |
+
"csa_dense_mode": false,
|
| 393 |
+
"use_fused_mhc": false,
|
| 394 |
+
"mhc_recompute_layer_num": null,
|
| 395 |
+
"megatron_extra_kwargs": null,
|
| 396 |
+
"language_model_only": false,
|
| 397 |
+
"check_model": true,
|
| 398 |
+
"apply_wd_to_qk_layernorm": false,
|
| 399 |
+
"linear_decoupled_in_proj": false,
|
| 400 |
+
"enable_dft_loss": false,
|
| 401 |
+
"enable_channel_loss": false,
|
| 402 |
+
"mrl_dims": null,
|
| 403 |
+
"save_strategy": "epoch",
|
| 404 |
+
"callbacks": [
|
| 405 |
+
"print",
|
| 406 |
+
"default_flow",
|
| 407 |
+
"swanlab"
|
| 408 |
+
],
|
| 409 |
+
"skip_megatron_init": false,
|
| 410 |
+
"add_version": false,
|
| 411 |
+
"create_checkpoint_symlink": false,
|
| 412 |
+
"swift_version": "4.4.0",
|
| 413 |
+
"ckpt_dir": null,
|
| 414 |
+
"rank": 0,
|
| 415 |
+
"global_world_size": 8,
|
| 416 |
+
"local_world_size": 8,
|
| 417 |
+
"model_suffix": "Qwen3-4B-Thinking-2507",
|
| 418 |
+
"model_info": "ModelInfo(model_type='qwen3', model_dir='/mnt/exp/moe-lite/tangzinan.tzn/models/Qwen3-4B-Thinking-2507', torch_dtype=torch.bfloat16, max_model_len=262144, quant_method=None, quant_bits=None, rope_scaling={'rope_theta': 5000000, 'rope_type': 'default'}, is_moe_model=False, is_multimodal=False, config=None, task_type='causal_lm', num_labels=None)",
|
| 419 |
+
"model_meta": "ModelMeta(model_type='qwen3', model_groups=[ModelGroup(models=[Model(ms_model_id='Qwen/Qwen3-0.6B-Base', hf_model_id='Qwen/Qwen3-0.6B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-1.7B-Base', hf_model_id='Qwen/Qwen3-1.7B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B-Base', hf_model_id='Qwen/Qwen3-4B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-8B-Base', hf_model_id='Qwen/Qwen3-8B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-14B-Base', hf_model_id='Qwen/Qwen3-14B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-0.6B', hf_model_id='Qwen/Qwen3-0.6B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-1.7B', hf_model_id='Qwen/Qwen3-1.7B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B', hf_model_id='Qwen/Qwen3-4B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-8B', hf_model_id='Qwen/Qwen3-8B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-14B', hf_model_id='Qwen/Qwen3-14B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-32B', hf_model_id='Qwen/Qwen3-32B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-0.6B-FP8', hf_model_id='Qwen/Qwen3-0.6B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-1.7B-FP8', hf_model_id='Qwen/Qwen3-1.7B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B-FP8', hf_model_id='Qwen/Qwen3-4B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-8B-FP8', hf_model_id='Qwen/Qwen3-8B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-14B-FP8', hf_model_id='Qwen/Qwen3-14B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-32B-FP8', hf_model_id='Qwen/Qwen3-32B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B-AWQ', hf_model_id='Qwen/Qwen3-4B-AWQ', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-8B-AWQ', hf_model_id='Qwen/Qwen3-8B-AWQ', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-14B-AWQ', hf_model_id='Qwen/Qwen3-14B-AWQ', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-32B-AWQ', hf_model_id='Qwen/Qwen3-32B-AWQ', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='swift/Qwen3-32B-AWQ', hf_model_id=None, model_path=None, ms_revision=None, hf_revision=None)], template='qwen3', ignore_patterns=None, requires=None, tags=[]), ModelGroup(models=[Model(ms_model_id='deepseek-ai/DeepSeek-R1-0528-Qwen3-8B', hf_model_id='deepseek-ai/DeepSeek-R1-0528-Qwen3-8B', model_path=None, ms_revision=None, hf_revision=None)], template='deepseek_r1', ignore_patterns=None, requires=None, tags=[]), ModelGroup(models=[Model(ms_model_id='Qwen/Qwen3Guard-Gen-0.6B', hf_model_id='Qwen/Qwen3Guard-Gen-0.6B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3Guard-Gen-4B', hf_model_id='Qwen/Qwen3Guard-Gen-4B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3Guard-Gen-8B', hf_model_id='Qwen/Qwen3Guard-Gen-8B', model_path=None, ms_revision=None, hf_revision=None)], template='qwen3_guard', ignore_patterns=None, requires=None, tags=[]), ModelGroup(models=[Model(ms_model_id='Alibaba-AAIG/YuFeng-XGuard-Reason-0.6B', hf_model_id='Alibaba-AAIG/YuFeng-XGuard-Reason-0.6B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Alibaba-AAIG/YuFeng-XGuard-Reason-8B', hf_model_id='Alibaba-AAIG/YuFeng-XGuard-Reason-8B', model_path=None, ms_revision=None, hf_revision=None)], template='yufeng_xguard', ignore_patterns=None, requires=None, tags=[]), ModelGroup(models=[Model(ms_model_id='Qwen/Qwen3-4B-Thinking-2507', hf_model_id='Qwen/Qwen3-4B-Thinking-2507', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B-Thinking-2507-FP8', hf_model_id='Qwen/Qwen3-4B-Thinking-2507-FP8', model_path=None, ms_revision=None, hf_revision=None)], template='qwen3_thinking', ignore_patterns=None, requires=None, tags=[]), ModelGroup(models=[Model(ms_model_id='Qwen/Qwen3-4B-Instruct-2507', hf_model_id='Qwen/Qwen3-4B-Instruct-2507', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3-4B-Instruct-2507-FP8', hf_model_id='Qwen/Qwen3-4B-Instruct-2507-FP8', model_path=None, ms_revision=None, hf_revision=None)], template='qwen3_nothinking', ignore_patterns=None, requires=None, tags=[])], loader=<class 'swift.model.register.ModelLoader'>, template='qwen3_thinking', model_arch=ModelKeys(arch_name='llama', embedding='model.embed_tokens', module_list='model.layers', lm_head='lm_head', q_proj='model.layers.{}.self_attn.q_proj', k_proj='model.layers.{}.self_attn.k_proj', v_proj='model.layers.{}.self_attn.v_proj', o_proj='model.layers.{}.self_attn.o_proj', attention='model.layers.{}.self_attn', mlp='model.layers.{}.mlp', down_proj='model.layers.{}.mlp.down_proj', qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None), mcore_model_type=None, architectures=['Qwen3ForCausalLM'], additional_saved_files=[], torch_dtype=None, is_multimodal=False, is_reward=False, task_type=None, ignore_patterns=None, requires=['transformers>=4.51'], tags=[])",
|
| 420 |
+
"model_dir": "/mnt/exp/moe-lite/tangzinan.tzn/models/Qwen3-4B-Thinking-2507",
|
| 421 |
+
"template_meta": "QwenTemplateMeta(template_type='qwen3_thinking', prefix=[], prompt=['<|im_start|>user\\n{{QUERY}}<|im_end|>\\n<|im_start|>assistant\\n'], chat_sep=['<|im_end|>\\n'], suffix=['<|im_end|>\\n'], template_cls=<class 'swift.template.base.Template'>, system_prefix=['<|im_start|>system\\n{{SYSTEM}}<|im_end|>\\n'], default_system=None, auto_add_bos=False, stop_words=['<|endoftext|>'], agent_template='hermes', is_thinking=True, thinking_prefix='<think>\\n', non_thinking_prefix='', history_thinking_prefix='')",
|
| 422 |
+
"_val_dataset_exists": false,
|
| 423 |
+
"hub": "<class 'swift.hub.hub.MSHub'>",
|
| 424 |
+
"seq_length": 65536,
|
| 425 |
+
"is_multimodal": false,
|
| 426 |
+
"megatron_model_meta": "ModelMeta(model_type='gpt', model_types=['qwen2', 'llama', 'qwen3', 'qwen2_moe', 'qwen3_moe', 'internlm3', 'mimo', 'deepseek', 'deepseek_v2', 'deepseek_v3', 'deepseek_v32', 'kimi_k2', 'dots1', 'ernie4_5', 'ernie4_5_moe', 'glm4_moe', 'glm4_moe_lite', 'glm_moe_dsa', 'gpt_oss'], bridge_cls=<class 'mcore_bridge.bridge.gpt_bridge.GPTBridge'>, visual_cls=None, is_multimodal=False, loader=<class 'mcore_bridge.model.register.ModelLoader'>)",
|
| 427 |
+
"fp8": null,
|
| 428 |
+
"fp4": null,
|
| 429 |
+
"world_size": 8,
|
| 430 |
+
"data_parallel_size": 2,
|
| 431 |
+
"num_microbatches": 8
|
| 432 |
+
}
|
assets/main.png
ADDED
|
Git LFS Details
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{{- messages[0].content + '\n\n' }}
|
| 5 |
+
{%- endif %}
|
| 6 |
+
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 7 |
+
{%- for tool in tools %}
|
| 8 |
+
{{- "\n" }}
|
| 9 |
+
{{- tool | tojson }}
|
| 10 |
+
{%- endfor %}
|
| 11 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 12 |
+
{%- else %}
|
| 13 |
+
{%- if messages[0].role == 'system' %}
|
| 14 |
+
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
| 15 |
+
{%- endif %}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 18 |
+
{%- for message in messages[::-1] %}
|
| 19 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 20 |
+
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
| 21 |
+
{%- set ns.multi_step_tool = false %}
|
| 22 |
+
{%- set ns.last_query_index = index %}
|
| 23 |
+
{%- endif %}
|
| 24 |
+
{%- endfor %}
|
| 25 |
+
{%- for message in messages %}
|
| 26 |
+
{%- if message.content is string %}
|
| 27 |
+
{%- set content = message.content %}
|
| 28 |
+
{%- else %}
|
| 29 |
+
{%- set content = '' %}
|
| 30 |
+
{%- endif %}
|
| 31 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 32 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 33 |
+
{%- elif message.role == "assistant" %}
|
| 34 |
+
{%- set reasoning_content = '' %}
|
| 35 |
+
{%- if message.reasoning_content is string %}
|
| 36 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 37 |
+
{%- else %}
|
| 38 |
+
{%- if '</think>' in content %}
|
| 39 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 40 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 41 |
+
{%- endif %}
|
| 42 |
+
{%- endif %}
|
| 43 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 44 |
+
{%- if loop.last or (not loop.last and reasoning_content) %}
|
| 45 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
| 46 |
+
{%- else %}
|
| 47 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 48 |
+
{%- endif %}
|
| 49 |
+
{%- else %}
|
| 50 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 51 |
+
{%- endif %}
|
| 52 |
+
{%- if message.tool_calls %}
|
| 53 |
+
{%- for tool_call in message.tool_calls %}
|
| 54 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 55 |
+
{{- '\n' }}
|
| 56 |
+
{%- endif %}
|
| 57 |
+
{%- if tool_call.function %}
|
| 58 |
+
{%- set tool_call = tool_call.function %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 61 |
+
{{- tool_call.name }}
|
| 62 |
+
{{- '", "arguments": ' }}
|
| 63 |
+
{%- if tool_call.arguments is string %}
|
| 64 |
+
{{- tool_call.arguments }}
|
| 65 |
+
{%- else %}
|
| 66 |
+
{{- tool_call.arguments | tojson }}
|
| 67 |
+
{%- endif %}
|
| 68 |
+
{{- '}\n</tool_call>' }}
|
| 69 |
+
{%- endfor %}
|
| 70 |
+
{%- endif %}
|
| 71 |
+
{{- '<|im_end|>\n' }}
|
| 72 |
+
{%- elif message.role == "tool" %}
|
| 73 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 74 |
+
{{- '<|im_start|>user' }}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{{- '\n<tool_response>\n' }}
|
| 77 |
+
{{- content }}
|
| 78 |
+
{{- '\n</tool_response>' }}
|
| 79 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 80 |
+
{{- '<|im_end|>\n' }}
|
| 81 |
+
{%- endif %}
|
| 82 |
+
{%- endif %}
|
| 83 |
+
{%- endfor %}
|
| 84 |
+
{%- if add_generation_prompt %}
|
| 85 |
+
{{- '<|im_start|>assistant\n<think>\n' }}
|
| 86 |
+
{%- endif %}
|
config.json
ADDED
|
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 151645,
|
| 10 |
+
"head_dim": 128,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2560,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 9728,
|
| 15 |
+
"layer_types": [
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"full_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"full_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"full_attention",
|
| 50 |
+
"full_attention",
|
| 51 |
+
"full_attention"
|
| 52 |
+
],
|
| 53 |
+
"max_position_embeddings": 262144,
|
| 54 |
+
"max_window_layers": 36,
|
| 55 |
+
"model_type": "qwen3",
|
| 56 |
+
"num_attention_heads": 32,
|
| 57 |
+
"num_hidden_layers": 36,
|
| 58 |
+
"num_key_value_heads": 8,
|
| 59 |
+
"pad_token_id": null,
|
| 60 |
+
"rms_norm_eps": 1e-06,
|
| 61 |
+
"rope_parameters": {
|
| 62 |
+
"rope_theta": 5000000,
|
| 63 |
+
"rope_type": "default"
|
| 64 |
+
},
|
| 65 |
+
"sliding_window": null,
|
| 66 |
+
"tie_word_embeddings": true,
|
| 67 |
+
"transformers_version": "5.12.1",
|
| 68 |
+
"use_cache": true,
|
| 69 |
+
"use_sliding_window": false,
|
| 70 |
+
"vocab_size": 151936
|
| 71 |
+
}
|
latest_checkpointed_iteration.txt
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
616
|
model-00001-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d3cbcf8f8cae6da0429d1e536599ffc00d53d181fb59852189a4cdab97d5518
|
| 3 |
+
size 4967220560
|
model-00002-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1eef0ae42f91bfa73c9339068ab6792a70e3d88d047785680c70051f5865f3cf
|
| 3 |
+
size 3077761384
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,405 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 8044936192
|
| 4 |
+
},
|
| 5 |
+
"weight_map": {
|
| 6 |
+
"model.embed_tokens.weight": "model-00001-of-00002.safetensors",
|
| 7 |
+
"model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 8 |
+
"model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 9 |
+
"model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 10 |
+
"model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 11 |
+
"model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 12 |
+
"model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 13 |
+
"model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 14 |
+
"model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 15 |
+
"model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 16 |
+
"model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 17 |
+
"model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 18 |
+
"model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 19 |
+
"model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 20 |
+
"model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 21 |
+
"model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 22 |
+
"model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 23 |
+
"model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 24 |
+
"model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 25 |
+
"model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 26 |
+
"model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 27 |
+
"model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 28 |
+
"model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 29 |
+
"model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 30 |
+
"model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 31 |
+
"model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 32 |
+
"model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 33 |
+
"model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 34 |
+
"model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 35 |
+
"model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 36 |
+
"model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 37 |
+
"model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 38 |
+
"model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 39 |
+
"model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 40 |
+
"model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 41 |
+
"model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 42 |
+
"model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 43 |
+
"model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 44 |
+
"model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 45 |
+
"model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 46 |
+
"model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 47 |
+
"model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 48 |
+
"model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 49 |
+
"model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 50 |
+
"model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 51 |
+
"model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 52 |
+
"model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 53 |
+
"model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 54 |
+
"model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 55 |
+
"model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 56 |
+
"model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 57 |
+
"model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 58 |
+
"model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 59 |
+
"model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 60 |
+
"model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 61 |
+
"model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 62 |
+
"model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 63 |
+
"model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 64 |
+
"model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 65 |
+
"model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 66 |
+
"model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 67 |
+
"model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 68 |
+
"model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 69 |
+
"model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 70 |
+
"model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 71 |
+
"model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 72 |
+
"model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 73 |
+
"model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 74 |
+
"model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 75 |
+
"model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 76 |
+
"model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 77 |
+
"model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 78 |
+
"model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 79 |
+
"model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 80 |
+
"model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 81 |
+
"model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 82 |
+
"model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 83 |
+
"model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 84 |
+
"model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 85 |
+
"model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 86 |
+
"model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 87 |
+
"model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 88 |
+
"model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 89 |
+
"model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 90 |
+
"model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 91 |
+
"model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 92 |
+
"model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 93 |
+
"model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 94 |
+
"model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 95 |
+
"model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 96 |
+
"model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 97 |
+
"model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 98 |
+
"model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 99 |
+
"model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 100 |
+
"model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 101 |
+
"model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 102 |
+
"model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 103 |
+
"model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 104 |
+
"model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 105 |
+
"model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 106 |
+
"model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 107 |
+
"model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 108 |
+
"model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 109 |
+
"model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 110 |
+
"model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 111 |
+
"model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 112 |
+
"model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 113 |
+
"model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 114 |
+
"model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 115 |
+
"model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 116 |
+
"model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 117 |
+
"model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 118 |
+
"model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 119 |
+
"model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 120 |
+
"model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 121 |
+
"model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 122 |
+
"model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 123 |
+
"model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 124 |
+
"model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 125 |
+
"model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 126 |
+
"model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 127 |
+
"model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 128 |
+
"model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 129 |
+
"model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 130 |
+
"model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 131 |
+
"model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 132 |
+
"model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 133 |
+
"model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 134 |
+
"model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 135 |
+
"model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 136 |
+
"model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 137 |
+
"model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 138 |
+
"model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 139 |
+
"model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 140 |
+
"model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 141 |
+
"model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 142 |
+
"model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 143 |
+
"model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 144 |
+
"model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 145 |
+
"model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 146 |
+
"model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 147 |
+
"model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 148 |
+
"model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 149 |
+
"model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 150 |
+
"model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 151 |
+
"model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 152 |
+
"model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 153 |
+
"model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 154 |
+
"model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 155 |
+
"model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 156 |
+
"model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 157 |
+
"model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 158 |
+
"model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 159 |
+
"model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 160 |
+
"model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 161 |
+
"model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 162 |
+
"model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 163 |
+
"model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 164 |
+
"model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 165 |
+
"model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 166 |
+
"model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 167 |
+
"model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 168 |
+
"model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 169 |
+
"model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 170 |
+
"model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 171 |
+
"model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 172 |
+
"model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 173 |
+
"model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 174 |
+
"model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 175 |
+
"model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 176 |
+
"model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 177 |
+
"model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 178 |
+
"model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 179 |
+
"model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 180 |
+
"model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 181 |
+
"model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 182 |
+
"model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 183 |
+
"model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 184 |
+
"model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 185 |
+
"model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 186 |
+
"model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 187 |
+
"model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 188 |
+
"model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 189 |
+
"model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 190 |
+
"model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 191 |
+
"model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 192 |
+
"model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 193 |
+
"model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 194 |
+
"model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 195 |
+
"model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 196 |
+
"model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 197 |
+
"model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 198 |
+
"model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 199 |
+
"model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 200 |
+
"model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 201 |
+
"model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 202 |
+
"model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 203 |
+
"model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 204 |
+
"model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 205 |
+
"model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 206 |
+
"model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 207 |
+
"model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 208 |
+
"model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 209 |
+
"model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 210 |
+
"model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 211 |
+
"model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 212 |
+
"model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 213 |
+
"model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 214 |
+
"model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 215 |
+
"model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 216 |
+
"model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 217 |
+
"model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 218 |
+
"model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 219 |
+
"model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 220 |
+
"model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 221 |
+
"model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 222 |
+
"model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 223 |
+
"model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 224 |
+
"model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 225 |
+
"model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 226 |
+
"model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 227 |
+
"model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 228 |
+
"model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 229 |
+
"model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 230 |
+
"model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 231 |
+
"model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 232 |
+
"model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 233 |
+
"model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 234 |
+
"model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 235 |
+
"model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 236 |
+
"model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 237 |
+
"model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 238 |
+
"model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 239 |
+
"model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 240 |
+
"model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 241 |
+
"model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 242 |
+
"model.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 243 |
+
"model.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 244 |
+
"model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 245 |
+
"model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 246 |
+
"model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 247 |
+
"model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 248 |
+
"model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 249 |
+
"model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 250 |
+
"model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 251 |
+
"model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 252 |
+
"model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 253 |
+
"model.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 254 |
+
"model.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 255 |
+
"model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 256 |
+
"model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 257 |
+
"model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 258 |
+
"model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 259 |
+
"model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 260 |
+
"model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 261 |
+
"model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 262 |
+
"model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 263 |
+
"model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 264 |
+
"model.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 265 |
+
"model.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 266 |
+
"model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 267 |
+
"model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 268 |
+
"model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 269 |
+
"model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 270 |
+
"model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 271 |
+
"model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 272 |
+
"model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 273 |
+
"model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 274 |
+
"model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 275 |
+
"model.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 276 |
+
"model.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 277 |
+
"model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 278 |
+
"model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 279 |
+
"model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 280 |
+
"model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 281 |
+
"model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 282 |
+
"model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 283 |
+
"model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 284 |
+
"model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 285 |
+
"model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 286 |
+
"model.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 287 |
+
"model.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 288 |
+
"model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 289 |
+
"model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 290 |
+
"model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 291 |
+
"model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 292 |
+
"model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 293 |
+
"model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 294 |
+
"model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 295 |
+
"model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 296 |
+
"model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 297 |
+
"model.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 298 |
+
"model.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 299 |
+
"model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 300 |
+
"model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 301 |
+
"model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 302 |
+
"model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 303 |
+
"model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 304 |
+
"model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 305 |
+
"model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 306 |
+
"model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 307 |
+
"model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 308 |
+
"model.layers.27.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 309 |
+
"model.layers.27.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 310 |
+
"model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 311 |
+
"model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 312 |
+
"model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 313 |
+
"model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 314 |
+
"model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 315 |
+
"model.layers.28.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 316 |
+
"model.layers.28.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 317 |
+
"model.layers.28.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 318 |
+
"model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 319 |
+
"model.layers.28.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 320 |
+
"model.layers.28.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 321 |
+
"model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 322 |
+
"model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 323 |
+
"model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 324 |
+
"model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 325 |
+
"model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 326 |
+
"model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 327 |
+
"model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 328 |
+
"model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 329 |
+
"model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 330 |
+
"model.layers.29.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 331 |
+
"model.layers.29.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 332 |
+
"model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 333 |
+
"model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 334 |
+
"model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 335 |
+
"model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 336 |
+
"model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 337 |
+
"model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 338 |
+
"model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 339 |
+
"model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 340 |
+
"model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 341 |
+
"model.layers.30.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 342 |
+
"model.layers.30.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 343 |
+
"model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 344 |
+
"model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 345 |
+
"model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 346 |
+
"model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 347 |
+
"model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 348 |
+
"model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 349 |
+
"model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 350 |
+
"model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 351 |
+
"model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 352 |
+
"model.layers.31.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 353 |
+
"model.layers.31.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 354 |
+
"model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 355 |
+
"model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 356 |
+
"model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 357 |
+
"model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 358 |
+
"model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 359 |
+
"model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 360 |
+
"model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 361 |
+
"model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 362 |
+
"model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 363 |
+
"model.layers.32.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 364 |
+
"model.layers.32.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 365 |
+
"model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 366 |
+
"model.layers.32.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 367 |
+
"model.layers.32.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 368 |
+
"model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 369 |
+
"model.layers.32.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 370 |
+
"model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 371 |
+
"model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 372 |
+
"model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 373 |
+
"model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 374 |
+
"model.layers.33.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 375 |
+
"model.layers.33.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 376 |
+
"model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 377 |
+
"model.layers.33.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 378 |
+
"model.layers.33.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 379 |
+
"model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 380 |
+
"model.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 381 |
+
"model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 382 |
+
"model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 383 |
+
"model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 384 |
+
"model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 385 |
+
"model.layers.34.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 386 |
+
"model.layers.34.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 387 |
+
"model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 388 |
+
"model.layers.34.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 389 |
+
"model.layers.34.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 390 |
+
"model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 391 |
+
"model.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 392 |
+
"model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 393 |
+
"model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 394 |
+
"model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 395 |
+
"model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 396 |
+
"model.layers.35.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 397 |
+
"model.layers.35.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 398 |
+
"model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 399 |
+
"model.layers.35.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 400 |
+
"model.layers.35.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 401 |
+
"model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 402 |
+
"model.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 403 |
+
"model.norm.weight": "model-00002-of-00002.safetensors"
|
| 404 |
+
}
|
| 405 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
| 3 |
+
size 11422650
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": null,
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"extra_special_tokens": [
|
| 9 |
+
"<|im_start|>",
|
| 10 |
+
"<|im_end|>",
|
| 11 |
+
"<|object_ref_start|>",
|
| 12 |
+
"<|object_ref_end|>",
|
| 13 |
+
"<|box_start|>",
|
| 14 |
+
"<|box_end|>",
|
| 15 |
+
"<|quad_start|>",
|
| 16 |
+
"<|quad_end|>",
|
| 17 |
+
"<|vision_start|>",
|
| 18 |
+
"<|vision_end|>",
|
| 19 |
+
"<|vision_pad|>",
|
| 20 |
+
"<|image_pad|>",
|
| 21 |
+
"<|video_pad|>"
|
| 22 |
+
],
|
| 23 |
+
"is_local": true,
|
| 24 |
+
"local_files_only": false,
|
| 25 |
+
"model_max_length": 262144,
|
| 26 |
+
"pad_token": "<|endoftext|>",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null
|
| 30 |
+
}
|