Add files using upload-large-folder tool
Browse files- README.md +114 -0
- config.json +64 -0
- config.official.yaml +141 -0
- mlx_manifest.json +64 -0
- model-00001-of-00024.safetensors +3 -0
- model-00002-of-00024.safetensors +3 -0
- model-00003-of-00024.safetensors +3 -0
- model-00004-of-00024.safetensors +3 -0
- model-00005-of-00024.safetensors +3 -0
- model-00006-of-00024.safetensors +3 -0
- model-00007-of-00024.safetensors +3 -0
- model-00008-of-00024.safetensors +3 -0
- model-00009-of-00024.safetensors +3 -0
- model-00010-of-00024.safetensors +3 -0
- model-00011-of-00024.safetensors +3 -0
- model-00012-of-00024.safetensors +3 -0
- model-00013-of-00024.safetensors +3 -0
- model-00014-of-00024.safetensors +3 -0
- model-00015-of-00024.safetensors +3 -0
- model-00016-of-00024.safetensors +3 -0
- model-00017-of-00024.safetensors +3 -0
- model-00018-of-00024.safetensors +3 -0
- model-00019-of-00024.safetensors +3 -0
- model-00020-of-00024.safetensors +3 -0
- model-00021-of-00024.safetensors +3 -0
- model-00022-of-00024.safetensors +3 -0
- model-00023-of-00024.safetensors +3 -0
- model-00024-of-00024.safetensors +3 -0
- model.safetensors.index.json +984 -0
- qwen2_tokenizer/config.json +27 -0
- qwen2_tokenizer/generation_config.json +7 -0
- qwen2_tokenizer/merges.txt +0 -0
- qwen2_tokenizer/tokenizer.json +0 -0
- qwen2_tokenizer/tokenizer_config.json +40 -0
- qwen2_tokenizer/vocab.json +0 -0
- vocab.yaml +13 -0
README.md
ADDED
|
@@ -0,0 +1,114 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
library_name: mlx
|
| 4 |
+
pipeline_tag: text-to-audio
|
| 5 |
+
base_model:
|
| 6 |
+
- tencent/SongGeneration
|
| 7 |
+
tags:
|
| 8 |
+
- mlx
|
| 9 |
+
- apple-silicon
|
| 10 |
+
- music-generation
|
| 11 |
+
- song-generation
|
| 12 |
+
- audiolm
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
Part of the SongGeneration MLX conversion set.
|
| 16 |
+
|
| 17 |
+
# SongGeneration-v2-medium-4bit
|
| 18 |
+
|
| 19 |
+
Apple MLX weights for the autoregressive `audiolm` token generator from Tencent SongGeneration v2-medium.
|
| 20 |
+
|
| 21 |
+
This is not a full-stack pure MLX audio pipeline yet: token generation runs with MLX, while FLAC decoding currently uses the official PyTorch Flow1dVAE / separate-tokenizer bridge in [`ailuntx/SongGeneration-MLX`](https://github.com/ailuntx/SongGeneration-MLX).
|
| 22 |
+
|
| 23 |
+
## TL;DR
|
| 24 |
+
|
| 25 |
+
| | |
|
| 26 |
+
|---|---|
|
| 27 |
+
| **Variant** | `v2-medium` |
|
| 28 |
+
| **Precision** | `4bit` |
|
| 29 |
+
| **Converted component** | SongGeneration `audiolm` token generator |
|
| 30 |
+
| **Runtime** | [`ailuntx/SongGeneration-MLX`](https://github.com/ailuntx/SongGeneration-MLX) |
|
| 31 |
+
| **Official model** | [`tencent/SongGeneration`](https://huggingface.co/tencent/SongGeneration) |
|
| 32 |
+
| **Official code** | [`tencent-ailab/songgeneration`](https://github.com/tencent-ailab/songgeneration) |
|
| 33 |
+
|
| 34 |
+
## Quick Start
|
| 35 |
+
|
| 36 |
+
```bash
|
| 37 |
+
hf download mlx-community/SongGeneration-v2-medium-4bit --local-dir ./models/SongGeneration-v2-medium-4bit
|
| 38 |
+
|
| 39 |
+
git clone https://github.com/ailuntx/SongGeneration-MLX.git
|
| 40 |
+
cd SongGeneration-MLX
|
| 41 |
+
python -m venv .venv
|
| 42 |
+
.venv/bin/pip install -e .
|
| 43 |
+
|
| 44 |
+
PYTHONPATH=. .venv/bin/python -m songgeneration_mlx.cli \
|
| 45 |
+
--model ./models/SongGeneration-v2-medium-4bit \
|
| 46 |
+
--lyrics '[verse] hello from mlx [chorus] sing it again' \
|
| 47 |
+
--description 'Pop, female vocal, bright production, [Musicality-medium].' \
|
| 48 |
+
--duration 2 \
|
| 49 |
+
--top-k 50 \
|
| 50 |
+
--temperature 0.9 \
|
| 51 |
+
--output tokens_2s.npz
|
| 52 |
+
```
|
| 53 |
+
|
| 54 |
+
To decode tokens to FLAC, use the official decoder bridge documented in the GitHub README. The bridge requires the official SongGeneration runtime assets.
|
| 55 |
+
|
| 56 |
+
## Variants
|
| 57 |
+
|
| 58 |
+
| Variant | Disk | Notes |
|
| 59 |
+
|---|---:|---|
|
| 60 |
+
| `SongGeneration-v2-medium-fp32` | 10G | high-precision medium baseline |
|
| 61 |
+
| `SongGeneration-v2-medium-bfloat16` | 5.2G | recommended medium quality baseline |
|
| 62 |
+
| `SongGeneration-v2-medium-8bit` | 2.8G | smaller medium checkpoint |
|
| 63 |
+
| `SongGeneration-v2-medium-4bit` | 1.5G | smallest medium checkpoint |
|
| 64 |
+
| `SongGeneration-v2-large-fp32` | 19G | high-precision large baseline |
|
| 65 |
+
| `SongGeneration-v2-large-bfloat16` | 9.5G | large quality baseline |
|
| 66 |
+
| `SongGeneration-v2-large-8bit` | 5.0G | smaller large checkpoint |
|
| 67 |
+
| `SongGeneration-v2-large-4bit` | 2.7G | smallest large checkpoint |
|
| 68 |
+
|
| 69 |
+
## Layout
|
| 70 |
+
|
| 71 |
+
```text
|
| 72 |
+
SongGeneration-v2-medium-4bit/
|
| 73 |
+
|-- model-00001-of-000xx.safetensors
|
| 74 |
+
|-- model.safetensors.index.json
|
| 75 |
+
|-- config.json
|
| 76 |
+
|-- mlx_manifest.json
|
| 77 |
+
|-- config.official.yaml
|
| 78 |
+
|-- vocab.yaml
|
| 79 |
+
`-- qwen2_tokenizer/
|
| 80 |
+
```
|
| 81 |
+
|
| 82 |
+
## Validation
|
| 83 |
+
|
| 84 |
+
Local Apple Silicon validation was run on the medium bfloat16 path:
|
| 85 |
+
|
| 86 |
+
| Test | Result |
|
| 87 |
+
|---|---|
|
| 88 |
+
| 12s MLX token generation | 550 pattern steps, about 1 minute wall time |
|
| 89 |
+
| 12s official decoder bridge | 73.27s wall time |
|
| 90 |
+
| 12s FLAC | 48kHz stereo, 12.000s, RMS about `0.163` |
|
| 91 |
+
|
| 92 |
+
The official recent-token repetition penalty is implemented in the MLX runtime. Without it, 12s generations collapse into repeated tokens and decode close to silence.
|
| 93 |
+
|
| 94 |
+
## License
|
| 95 |
+
|
| 96 |
+
License follows the upstream SongGeneration release. Check the official model card and repository for the authoritative model license.
|
| 97 |
+
|
| 98 |
+
## Citation
|
| 99 |
+
|
| 100 |
+
```bibtex
|
| 101 |
+
@misc{songgeneration-mlx,
|
| 102 |
+
title = {SongGeneration-MLX: Apple MLX port of SongGeneration},
|
| 103 |
+
author = {ailuntx},
|
| 104 |
+
year = {2026},
|
| 105 |
+
url = {https://github.com/ailuntx/SongGeneration-MLX},
|
| 106 |
+
}
|
| 107 |
+
|
| 108 |
+
@article{lei2025levo,
|
| 109 |
+
title = {LeVo: High-Quality Song Generation with Multi-Preference Alignment},
|
| 110 |
+
author = {Lei, Shun and Xu, Yaoxun and Lin, Zhiwei and Zhang, Huaicheng and Tan, Wei and Chen, Hangting and Yu, Jianwei and Zhang, Yixuan and Yang, Chenyu and Zhu, Haina and Wang, Shuai and Wu, Zhiyong and Yu, Dong},
|
| 111 |
+
journal = {arXiv preprint arXiv:2506.07520},
|
| 112 |
+
year = {2025},
|
| 113 |
+
}
|
| 114 |
+
```
|
config.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"format": "songgeneration-mlx-audiolm",
|
| 3 |
+
"source": "tencent/SongGeneration",
|
| 4 |
+
"official_code": "https://github.com/tencent-ailab/songgeneration",
|
| 5 |
+
"variant": "v2-medium",
|
| 6 |
+
"precision": "4bit",
|
| 7 |
+
"runtime": {
|
| 8 |
+
"hidden_size": 1536,
|
| 9 |
+
"intermediate_size": 8960,
|
| 10 |
+
"num_heads": 12,
|
| 11 |
+
"num_layers": 28,
|
| 12 |
+
"num_layers_sub": 12,
|
| 13 |
+
"code_depth": 3,
|
| 14 |
+
"code_size": 16384,
|
| 15 |
+
"prompt_len": 10,
|
| 16 |
+
"frame_rate": 25,
|
| 17 |
+
"max_position_embeddings": 10000,
|
| 18 |
+
"max_position_embeddings_sub": 10000,
|
| 19 |
+
"rope_theta": 500000.0,
|
| 20 |
+
"rope_theta_sub": 500000.0,
|
| 21 |
+
"cfg_coef": 1.5
|
| 22 |
+
},
|
| 23 |
+
"components": {
|
| 24 |
+
"audiolm": "converted to MLX safetensors",
|
| 25 |
+
"qwen_tokenizer": "copied for lyric/style tokenization",
|
| 26 |
+
"audio_decoder": "use official PyTorch Flow1dVAE/separate tokenizer bridge for now"
|
| 27 |
+
},
|
| 28 |
+
"skipped_keys": [
|
| 29 |
+
"audiolm.out_norm.weight",
|
| 30 |
+
"audiolm.out_norm.bias",
|
| 31 |
+
"audiolm.transformer2.lm_head.weight"
|
| 32 |
+
],
|
| 33 |
+
"quantization": {
|
| 34 |
+
"bits": 4,
|
| 35 |
+
"group_size": 64,
|
| 36 |
+
"mode": "affine"
|
| 37 |
+
},
|
| 38 |
+
"weight_files": [
|
| 39 |
+
"model-00001-of-00024.safetensors",
|
| 40 |
+
"model-00002-of-00024.safetensors",
|
| 41 |
+
"model-00003-of-00024.safetensors",
|
| 42 |
+
"model-00004-of-00024.safetensors",
|
| 43 |
+
"model-00005-of-00024.safetensors",
|
| 44 |
+
"model-00006-of-00024.safetensors",
|
| 45 |
+
"model-00007-of-00024.safetensors",
|
| 46 |
+
"model-00008-of-00024.safetensors",
|
| 47 |
+
"model-00009-of-00024.safetensors",
|
| 48 |
+
"model-00010-of-00024.safetensors",
|
| 49 |
+
"model-00011-of-00024.safetensors",
|
| 50 |
+
"model-00012-of-00024.safetensors",
|
| 51 |
+
"model-00013-of-00024.safetensors",
|
| 52 |
+
"model-00014-of-00024.safetensors",
|
| 53 |
+
"model-00015-of-00024.safetensors",
|
| 54 |
+
"model-00016-of-00024.safetensors",
|
| 55 |
+
"model-00017-of-00024.safetensors",
|
| 56 |
+
"model-00018-of-00024.safetensors",
|
| 57 |
+
"model-00019-of-00024.safetensors",
|
| 58 |
+
"model-00020-of-00024.safetensors",
|
| 59 |
+
"model-00021-of-00024.safetensors",
|
| 60 |
+
"model-00022-of-00024.safetensors",
|
| 61 |
+
"model-00023-of-00024.safetensors",
|
| 62 |
+
"model-00024-of-00024.safetensors"
|
| 63 |
+
]
|
| 64 |
+
}
|
config.official.yaml
ADDED
|
@@ -0,0 +1,141 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# ================ Train Config ================ #
|
| 2 |
+
lyric_processor:
|
| 3 |
+
max_dur: 270
|
| 4 |
+
min_dur: 30
|
| 5 |
+
prompt_len: 10
|
| 6 |
+
pad_to_max: true
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
# ================ Audio tokenzier ================ #
|
| 10 |
+
audio_tokenizer_checkpoint: Flow1dVAE1rvq_./ckpt/model_1rvq/model_2_fixed.safetensors
|
| 11 |
+
audio_tokenizer_frame_rate: 25
|
| 12 |
+
audio_tokenizer_code_depth: 1
|
| 13 |
+
sample_rate: 48000
|
| 14 |
+
|
| 15 |
+
audio_tokenizer_checkpoint_sep: Flow1dVAESeparate_./ckpt/model_septoken/model_2.safetensors
|
| 16 |
+
audio_tokenizer_frame_rate_sep: 25
|
| 17 |
+
audio_tokenizer_code_depth_sep: 2
|
| 18 |
+
sample_rate_sep: 48000
|
| 19 |
+
|
| 20 |
+
# ================ VAE ================ #
|
| 21 |
+
vae_config: ./ckpt/vae/stable_audio_1920_vae.json
|
| 22 |
+
vae_model: ./ckpt/vae/autoencoder_music_1320k.ckpt
|
| 23 |
+
|
| 24 |
+
# ================== LM =========================== #
|
| 25 |
+
lm:
|
| 26 |
+
lm_type: Llama # [Llama]
|
| 27 |
+
dim: 1536
|
| 28 |
+
intermediate_size: 8960
|
| 29 |
+
num_heads: 12
|
| 30 |
+
num_layers: 28
|
| 31 |
+
num_layers_sub: 12
|
| 32 |
+
code_depth: 3
|
| 33 |
+
code_size: 16384
|
| 34 |
+
max_position_embeddings: 10000
|
| 35 |
+
max_position_embeddings_sub: 10000
|
| 36 |
+
rope_theta: 500000.0
|
| 37 |
+
rope_theta_sub: 500000.0
|
| 38 |
+
dropout: 0.0
|
| 39 |
+
use_flash_attn_2: true
|
| 40 |
+
activation: gelu
|
| 41 |
+
norm_first: true
|
| 42 |
+
bias_ff: false
|
| 43 |
+
bias_attn: false
|
| 44 |
+
causal: true
|
| 45 |
+
custom: false
|
| 46 |
+
memory_efficient: true
|
| 47 |
+
attention_as_float32: false
|
| 48 |
+
layer_scale: null
|
| 49 |
+
positional_embedding: sin
|
| 50 |
+
xpos: false
|
| 51 |
+
checkpointing: torch
|
| 52 |
+
weight_init: gaussian
|
| 53 |
+
depthwise_init: current
|
| 54 |
+
zero_bias_init: true
|
| 55 |
+
norm: layer_norm
|
| 56 |
+
cross_attention: false
|
| 57 |
+
qk_layer_norm: false
|
| 58 |
+
qk_layer_norm_cross: false
|
| 59 |
+
attention_dropout: null
|
| 60 |
+
kv_repeat: 1
|
| 61 |
+
|
| 62 |
+
codebooks_pattern:
|
| 63 |
+
modeling: delay
|
| 64 |
+
delay:
|
| 65 |
+
delays: [ 0, 250, 250 ]
|
| 66 |
+
flatten_first: 0
|
| 67 |
+
empty_initial: 0
|
| 68 |
+
|
| 69 |
+
# ================ Conditioners ===================== #
|
| 70 |
+
classifier_free_guidance:
|
| 71 |
+
# drop all conditions simultaneously
|
| 72 |
+
training_dropout: 0.15
|
| 73 |
+
inference_coef: 1.5
|
| 74 |
+
|
| 75 |
+
attribute_dropout:
|
| 76 |
+
# drop each condition separately
|
| 77 |
+
args:
|
| 78 |
+
active_on_eval: false
|
| 79 |
+
text:
|
| 80 |
+
description: 0.0
|
| 81 |
+
type_info: 0.2
|
| 82 |
+
audio:
|
| 83 |
+
prompt_audio: 0.5
|
| 84 |
+
|
| 85 |
+
|
| 86 |
+
use_text_training: True
|
| 87 |
+
fuser:
|
| 88 |
+
sum: []
|
| 89 |
+
prepend: [ description, prompt_audio, type_info ] # this order is the SAME with the input concatenation order
|
| 90 |
+
|
| 91 |
+
conditioners:
|
| 92 |
+
prompt_audio:
|
| 93 |
+
model: qt_embedding
|
| 94 |
+
qt_embedding:
|
| 95 |
+
code_size: 16384
|
| 96 |
+
code_depth: 3
|
| 97 |
+
max_len: ${eval:${prompt_len}*${audio_tokenizer_frame_rate}+2} # 25*10+2+1
|
| 98 |
+
description:
|
| 99 |
+
model: QwTokenizer
|
| 100 |
+
QwTokenizer:
|
| 101 |
+
token_path: third_party/Qwen2-7B
|
| 102 |
+
max_len: 600
|
| 103 |
+
add_token_list: ${load_yaml:conf/vocab.yaml}
|
| 104 |
+
type_info:
|
| 105 |
+
model: QwTextTokenizer
|
| 106 |
+
QwTextTokenizer:
|
| 107 |
+
token_path: third_party/Qwen2-7B
|
| 108 |
+
max_len: 100
|
| 109 |
+
|
| 110 |
+
offload:
|
| 111 |
+
audiolm:
|
| 112 |
+
offload_module: self
|
| 113 |
+
cpu_mem_gb: 0
|
| 114 |
+
pre_copy_step: 1
|
| 115 |
+
clean_cache_after_forward: false
|
| 116 |
+
dtype: torch.float16
|
| 117 |
+
offload_layer_dict:
|
| 118 |
+
transformer: 4
|
| 119 |
+
transformer2: 4
|
| 120 |
+
ignore_layer_list: []
|
| 121 |
+
clean_cache_wrapper:
|
| 122 |
+
module: self
|
| 123 |
+
method_name: _sample_next_token
|
| 124 |
+
diff_mem_gb_thre: 2
|
| 125 |
+
debug: false
|
| 126 |
+
|
| 127 |
+
wav_tokenizer_diffusion:
|
| 128 |
+
offload_module: self.model.model
|
| 129 |
+
pre_copy_step: 1
|
| 130 |
+
clean_cache_after_forward: false
|
| 131 |
+
cpu_mem_gb: -1
|
| 132 |
+
dtype: null
|
| 133 |
+
offload_layer_dict:
|
| 134 |
+
cfm_wrapper: 5
|
| 135 |
+
hubert: 4
|
| 136 |
+
ignore_layer_list: []
|
| 137 |
+
clean_cache_wrapper:
|
| 138 |
+
module: self.model.model.cfm_wrapper.estimator
|
| 139 |
+
method_name: forward
|
| 140 |
+
diff_mem_gb_thre: 1
|
| 141 |
+
debug: false
|
mlx_manifest.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"format": "songgeneration-mlx-audiolm",
|
| 3 |
+
"source": "tencent/SongGeneration",
|
| 4 |
+
"official_code": "https://github.com/tencent-ailab/songgeneration",
|
| 5 |
+
"variant": "v2-medium",
|
| 6 |
+
"precision": "4bit",
|
| 7 |
+
"runtime": {
|
| 8 |
+
"hidden_size": 1536,
|
| 9 |
+
"intermediate_size": 8960,
|
| 10 |
+
"num_heads": 12,
|
| 11 |
+
"num_layers": 28,
|
| 12 |
+
"num_layers_sub": 12,
|
| 13 |
+
"code_depth": 3,
|
| 14 |
+
"code_size": 16384,
|
| 15 |
+
"prompt_len": 10,
|
| 16 |
+
"frame_rate": 25,
|
| 17 |
+
"max_position_embeddings": 10000,
|
| 18 |
+
"max_position_embeddings_sub": 10000,
|
| 19 |
+
"rope_theta": 500000.0,
|
| 20 |
+
"rope_theta_sub": 500000.0,
|
| 21 |
+
"cfg_coef": 1.5
|
| 22 |
+
},
|
| 23 |
+
"components": {
|
| 24 |
+
"audiolm": "converted to MLX safetensors",
|
| 25 |
+
"qwen_tokenizer": "copied for lyric/style tokenization",
|
| 26 |
+
"audio_decoder": "use official PyTorch Flow1dVAE/separate tokenizer bridge for now"
|
| 27 |
+
},
|
| 28 |
+
"skipped_keys": [
|
| 29 |
+
"audiolm.out_norm.weight",
|
| 30 |
+
"audiolm.out_norm.bias",
|
| 31 |
+
"audiolm.transformer2.lm_head.weight"
|
| 32 |
+
],
|
| 33 |
+
"quantization": {
|
| 34 |
+
"bits": 4,
|
| 35 |
+
"group_size": 64,
|
| 36 |
+
"mode": "affine"
|
| 37 |
+
},
|
| 38 |
+
"weight_files": [
|
| 39 |
+
"model-00001-of-00024.safetensors",
|
| 40 |
+
"model-00002-of-00024.safetensors",
|
| 41 |
+
"model-00003-of-00024.safetensors",
|
| 42 |
+
"model-00004-of-00024.safetensors",
|
| 43 |
+
"model-00005-of-00024.safetensors",
|
| 44 |
+
"model-00006-of-00024.safetensors",
|
| 45 |
+
"model-00007-of-00024.safetensors",
|
| 46 |
+
"model-00008-of-00024.safetensors",
|
| 47 |
+
"model-00009-of-00024.safetensors",
|
| 48 |
+
"model-00010-of-00024.safetensors",
|
| 49 |
+
"model-00011-of-00024.safetensors",
|
| 50 |
+
"model-00012-of-00024.safetensors",
|
| 51 |
+
"model-00013-of-00024.safetensors",
|
| 52 |
+
"model-00014-of-00024.safetensors",
|
| 53 |
+
"model-00015-of-00024.safetensors",
|
| 54 |
+
"model-00016-of-00024.safetensors",
|
| 55 |
+
"model-00017-of-00024.safetensors",
|
| 56 |
+
"model-00018-of-00024.safetensors",
|
| 57 |
+
"model-00019-of-00024.safetensors",
|
| 58 |
+
"model-00020-of-00024.safetensors",
|
| 59 |
+
"model-00021-of-00024.safetensors",
|
| 60 |
+
"model-00022-of-00024.safetensors",
|
| 61 |
+
"model-00023-of-00024.safetensors",
|
| 62 |
+
"model-00024-of-00024.safetensors"
|
| 63 |
+
]
|
| 64 |
+
}
|
model-00001-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8fc91af414b31fa862530a36c707a5c342a82eacf1dfb667ad7cc7a0b98049ed
|
| 3 |
+
size 14559480
|
model-00002-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e9cab73947c4bdb735e543065290f19cd60fd1523518c7a1796fbb2a8af5013
|
| 3 |
+
size 116474224
|
model-00003-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6dfe514be85f92267347d2dfeff183fc1d96d63a3d68a949038a75cc4524f85c
|
| 3 |
+
size 58377352
|
model-00004-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d08991a7a4aae48edb3806f1f79af45250425d219bb701f8a266d016015a022
|
| 3 |
+
size 60617704
|
model-00005-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:05355c61954554223b18c839dd4cd264da993150c56fb0bf78ae71a3ef6b0ec8
|
| 3 |
+
size 63894736
|
model-00006-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d65038727a45139c28d5ea08a93003f58822f2d868769d8c87ba9e7730402a31
|
| 3 |
+
size 64824896
|
model-00007-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4ffd0b32a23fc53291c54c751b258a97fe201af4325122d97af1c93bee1dc184
|
| 3 |
+
size 66770328
|
model-00008-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ed35d56205f74a321ed0d181c2afea0b167cdef8040b87eaa0ea21fc1bef2bdb
|
| 3 |
+
size 60453928
|
model-00009-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d917f8e3797bb70b8f187c1f67c2ee79abb826101ee6d7344b90a5202868c49d
|
| 3 |
+
size 64824920
|
model-00010-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28455144472660ab476354be97c1bf141e7b3f4754e58ef3052d6becdb67ae45
|
| 3 |
+
size 64824904
|
model-00011-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a98f34b65bb6807fc5353bc6811d119e4d802d0bce5d1a08cd319ea44a0d5bf
|
| 3 |
+
size 66770312
|
model-00012-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ddb3425d4f5ca110079255a89632f5fd2135c56f8145783511e8f1aa6076ceb
|
| 3 |
+
size 60453928
|
model-00013-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:59ab43548f038468e9fda3c0deeff2c82c046d465b3ffdcde9edf311d33dea9b
|
| 3 |
+
size 64824920
|
model-00014-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b117cf4c162e2186b7589c27c1c45013b839993c77ec74112dc54336babec062
|
| 3 |
+
size 64824904
|
model-00015-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a113cab19c645aec3a240b1a92f4be92cc3017289191e9e64a16b62f0fb3ffe7
|
| 3 |
+
size 66770272
|
model-00016-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:84d8c266bb0c6d53f1642699a4287d9206a4e9db624a0abdbaa11b11926af11e
|
| 3 |
+
size 60453880
|
model-00017-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:10bccf197ec87f76ae45b883e404bc00d0229b718164cc3eb82484eabf8eb429
|
| 3 |
+
size 57795848
|
model-00018-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:82a7296f417d76b6f54f588bce5c005264fa9d0500b076ea1cca92dee8852ccc
|
| 3 |
+
size 67085848
|
model-00019-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6c90acd63798fd1dfc16e5a6841b2fa535bcde51714ad683ddeb30eb15f6e992
|
| 3 |
+
size 60527824
|
model-00020-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d5a533b95f5e00c997fe14450eb93ee1463579cf1c01bb12a3ed9eb0a30bdc0
|
| 3 |
+
size 64824920
|
model-00021-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4596f5656191f0ab8c3d809ee0bc3ee72fde9b37fb23a0a669ea630639e2ad2e
|
| 3 |
+
size 64824912
|
model-00022-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8210bd3d1b7097bcfcf470548f36602197736d37c3f829268586648fc3e8d25
|
| 3 |
+
size 66770328
|
model-00023-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e6b0a6cac45aaadf251eac74e6c04ea6270062240275f784a1578e948129073d
|
| 3 |
+
size 59767696
|
model-00024-of-00024.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:05b164197bd01d9bb0c548e13523a3df5b1faa348ce5b9d66beae8a3ee7678e3
|
| 3 |
+
size 116468848
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,984 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 1577674176
|
| 4 |
+
},
|
| 5 |
+
"weight_map": {
|
| 6 |
+
"description_output_proj.biases": "model-00001-of-00024.safetensors",
|
| 7 |
+
"description_output_proj.scales": "model-00001-of-00024.safetensors",
|
| 8 |
+
"description_output_proj.weight": "model-00002-of-00024.safetensors",
|
| 9 |
+
"description_structure_emb.biases": "model-00003-of-00024.safetensors",
|
| 10 |
+
"description_structure_emb.scales": "model-00003-of-00024.safetensors",
|
| 11 |
+
"description_structure_emb.weight": "model-00003-of-00024.safetensors",
|
| 12 |
+
"emb0.biases": "model-00003-of-00024.safetensors",
|
| 13 |
+
"emb0.scales": "model-00003-of-00024.safetensors",
|
| 14 |
+
"emb0.weight": "model-00003-of-00024.safetensors",
|
| 15 |
+
"layer2_emb0.biases": "model-00003-of-00024.safetensors",
|
| 16 |
+
"layer2_emb0.scales": "model-00003-of-00024.safetensors",
|
| 17 |
+
"layer2_emb0.weight": "model-00003-of-00024.safetensors",
|
| 18 |
+
"layer2_emb1.biases": "model-00003-of-00024.safetensors",
|
| 19 |
+
"layer2_emb1.scales": "model-00003-of-00024.safetensors",
|
| 20 |
+
"layer2_emb1.weight": "model-00003-of-00024.safetensors",
|
| 21 |
+
"layer2_emb2.biases": "model-00003-of-00024.safetensors",
|
| 22 |
+
"layer2_emb2.scales": "model-00003-of-00024.safetensors",
|
| 23 |
+
"layer2_emb2.weight": "model-00003-of-00024.safetensors",
|
| 24 |
+
"linears0.biases": "model-00003-of-00024.safetensors",
|
| 25 |
+
"linears0.scales": "model-00003-of-00024.safetensors",
|
| 26 |
+
"linears0.weight": "model-00004-of-00024.safetensors",
|
| 27 |
+
"linears1.biases": "model-00004-of-00024.safetensors",
|
| 28 |
+
"linears1.scales": "model-00004-of-00024.safetensors",
|
| 29 |
+
"linears1.weight": "model-00004-of-00024.safetensors",
|
| 30 |
+
"mlp.linear0.bias": "model-00004-of-00024.safetensors",
|
| 31 |
+
"mlp.linear0.biases": "model-00004-of-00024.safetensors",
|
| 32 |
+
"mlp.linear0.scales": "model-00004-of-00024.safetensors",
|
| 33 |
+
"mlp.linear0.weight": "model-00004-of-00024.safetensors",
|
| 34 |
+
"mlp.linear2.bias": "model-00004-of-00024.safetensors",
|
| 35 |
+
"mlp.linear2.biases": "model-00004-of-00024.safetensors",
|
| 36 |
+
"mlp.linear2.scales": "model-00004-of-00024.safetensors",
|
| 37 |
+
"mlp.linear2.weight": "model-00004-of-00024.safetensors",
|
| 38 |
+
"prompt_audio_emb0.biases": "model-00004-of-00024.safetensors",
|
| 39 |
+
"prompt_audio_emb0.scales": "model-00004-of-00024.safetensors",
|
| 40 |
+
"prompt_audio_emb0.weight": "model-00004-of-00024.safetensors",
|
| 41 |
+
"prompt_audio_emb1.biases": "model-00004-of-00024.safetensors",
|
| 42 |
+
"prompt_audio_emb1.scales": "model-00004-of-00024.safetensors",
|
| 43 |
+
"prompt_audio_emb1.weight": "model-00004-of-00024.safetensors",
|
| 44 |
+
"prompt_audio_emb2.biases": "model-00004-of-00024.safetensors",
|
| 45 |
+
"prompt_audio_emb2.scales": "model-00004-of-00024.safetensors",
|
| 46 |
+
"prompt_audio_emb2.weight": "model-00005-of-00024.safetensors",
|
| 47 |
+
"prompt_audio_eot": "model-00005-of-00024.safetensors",
|
| 48 |
+
"prompt_audio_layer2_eot": "model-00005-of-00024.safetensors",
|
| 49 |
+
"transformer.embed_tokens.biases": "model-00005-of-00024.safetensors",
|
| 50 |
+
"transformer.embed_tokens.scales": "model-00005-of-00024.safetensors",
|
| 51 |
+
"transformer.embed_tokens.weight": "model-00005-of-00024.safetensors",
|
| 52 |
+
"transformer.layers.0.input_layernorm.weight": "model-00005-of-00024.safetensors",
|
| 53 |
+
"transformer.layers.0.mlp.down_proj.biases": "model-00005-of-00024.safetensors",
|
| 54 |
+
"transformer.layers.0.mlp.down_proj.scales": "model-00005-of-00024.safetensors",
|
| 55 |
+
"transformer.layers.0.mlp.down_proj.weight": "model-00005-of-00024.safetensors",
|
| 56 |
+
"transformer.layers.0.mlp.gate_proj.biases": "model-00005-of-00024.safetensors",
|
| 57 |
+
"transformer.layers.0.mlp.gate_proj.scales": "model-00005-of-00024.safetensors",
|
| 58 |
+
"transformer.layers.0.mlp.gate_proj.weight": "model-00005-of-00024.safetensors",
|
| 59 |
+
"transformer.layers.0.mlp.up_proj.biases": "model-00005-of-00024.safetensors",
|
| 60 |
+
"transformer.layers.0.mlp.up_proj.scales": "model-00005-of-00024.safetensors",
|
| 61 |
+
"transformer.layers.0.mlp.up_proj.weight": "model-00005-of-00024.safetensors",
|
| 62 |
+
"transformer.layers.0.post_attention_layernorm.weight": "model-00005-of-00024.safetensors",
|
| 63 |
+
"transformer.layers.0.self_attn.k_proj.biases": "model-00005-of-00024.safetensors",
|
| 64 |
+
"transformer.layers.0.self_attn.k_proj.scales": "model-00005-of-00024.safetensors",
|
| 65 |
+
"transformer.layers.0.self_attn.k_proj.weight": "model-00005-of-00024.safetensors",
|
| 66 |
+
"transformer.layers.0.self_attn.o_proj.biases": "model-00005-of-00024.safetensors",
|
| 67 |
+
"transformer.layers.0.self_attn.o_proj.scales": "model-00005-of-00024.safetensors",
|
| 68 |
+
"transformer.layers.0.self_attn.o_proj.weight": "model-00005-of-00024.safetensors",
|
| 69 |
+
"transformer.layers.0.self_attn.q_proj.biases": "model-00005-of-00024.safetensors",
|
| 70 |
+
"transformer.layers.0.self_attn.q_proj.scales": "model-00005-of-00024.safetensors",
|
| 71 |
+
"transformer.layers.0.self_attn.q_proj.weight": "model-00005-of-00024.safetensors",
|
| 72 |
+
"transformer.layers.0.self_attn.v_proj.biases": "model-00005-of-00024.safetensors",
|
| 73 |
+
"transformer.layers.0.self_attn.v_proj.scales": "model-00005-of-00024.safetensors",
|
| 74 |
+
"transformer.layers.0.self_attn.v_proj.weight": "model-00005-of-00024.safetensors",
|
| 75 |
+
"transformer.layers.1.input_layernorm.weight": "model-00005-of-00024.safetensors",
|
| 76 |
+
"transformer.layers.1.mlp.down_proj.biases": "model-00005-of-00024.safetensors",
|
| 77 |
+
"transformer.layers.1.mlp.down_proj.scales": "model-00005-of-00024.safetensors",
|
| 78 |
+
"transformer.layers.1.mlp.down_proj.weight": "model-00005-of-00024.safetensors",
|
| 79 |
+
"transformer.layers.1.mlp.gate_proj.biases": "model-00005-of-00024.safetensors",
|
| 80 |
+
"transformer.layers.1.mlp.gate_proj.scales": "model-00005-of-00024.safetensors",
|
| 81 |
+
"transformer.layers.1.mlp.gate_proj.weight": "model-00006-of-00024.safetensors",
|
| 82 |
+
"transformer.layers.1.mlp.up_proj.biases": "model-00006-of-00024.safetensors",
|
| 83 |
+
"transformer.layers.1.mlp.up_proj.scales": "model-00006-of-00024.safetensors",
|
| 84 |
+
"transformer.layers.1.mlp.up_proj.weight": "model-00006-of-00024.safetensors",
|
| 85 |
+
"transformer.layers.1.post_attention_layernorm.weight": "model-00006-of-00024.safetensors",
|
| 86 |
+
"transformer.layers.1.self_attn.k_proj.biases": "model-00006-of-00024.safetensors",
|
| 87 |
+
"transformer.layers.1.self_attn.k_proj.scales": "model-00006-of-00024.safetensors",
|
| 88 |
+
"transformer.layers.1.self_attn.k_proj.weight": "model-00006-of-00024.safetensors",
|
| 89 |
+
"transformer.layers.1.self_attn.o_proj.biases": "model-00006-of-00024.safetensors",
|
| 90 |
+
"transformer.layers.1.self_attn.o_proj.scales": "model-00006-of-00024.safetensors",
|
| 91 |
+
"transformer.layers.1.self_attn.o_proj.weight": "model-00006-of-00024.safetensors",
|
| 92 |
+
"transformer.layers.1.self_attn.q_proj.biases": "model-00006-of-00024.safetensors",
|
| 93 |
+
"transformer.layers.1.self_attn.q_proj.scales": "model-00006-of-00024.safetensors",
|
| 94 |
+
"transformer.layers.1.self_attn.q_proj.weight": "model-00006-of-00024.safetensors",
|
| 95 |
+
"transformer.layers.1.self_attn.v_proj.biases": "model-00006-of-00024.safetensors",
|
| 96 |
+
"transformer.layers.1.self_attn.v_proj.scales": "model-00006-of-00024.safetensors",
|
| 97 |
+
"transformer.layers.1.self_attn.v_proj.weight": "model-00006-of-00024.safetensors",
|
| 98 |
+
"transformer.layers.10.input_layernorm.weight": "model-00006-of-00024.safetensors",
|
| 99 |
+
"transformer.layers.10.mlp.down_proj.biases": "model-00006-of-00024.safetensors",
|
| 100 |
+
"transformer.layers.10.mlp.down_proj.scales": "model-00006-of-00024.safetensors",
|
| 101 |
+
"transformer.layers.10.mlp.down_proj.weight": "model-00006-of-00024.safetensors",
|
| 102 |
+
"transformer.layers.10.mlp.gate_proj.biases": "model-00006-of-00024.safetensors",
|
| 103 |
+
"transformer.layers.10.mlp.gate_proj.scales": "model-00006-of-00024.safetensors",
|
| 104 |
+
"transformer.layers.10.mlp.gate_proj.weight": "model-00006-of-00024.safetensors",
|
| 105 |
+
"transformer.layers.10.mlp.up_proj.biases": "model-00006-of-00024.safetensors",
|
| 106 |
+
"transformer.layers.10.mlp.up_proj.scales": "model-00006-of-00024.safetensors",
|
| 107 |
+
"transformer.layers.10.mlp.up_proj.weight": "model-00006-of-00024.safetensors",
|
| 108 |
+
"transformer.layers.10.post_attention_layernorm.weight": "model-00006-of-00024.safetensors",
|
| 109 |
+
"transformer.layers.10.self_attn.k_proj.biases": "model-00006-of-00024.safetensors",
|
| 110 |
+
"transformer.layers.10.self_attn.k_proj.scales": "model-00006-of-00024.safetensors",
|
| 111 |
+
"transformer.layers.10.self_attn.k_proj.weight": "model-00006-of-00024.safetensors",
|
| 112 |
+
"transformer.layers.10.self_attn.o_proj.biases": "model-00006-of-00024.safetensors",
|
| 113 |
+
"transformer.layers.10.self_attn.o_proj.scales": "model-00006-of-00024.safetensors",
|
| 114 |
+
"transformer.layers.10.self_attn.o_proj.weight": "model-00006-of-00024.safetensors",
|
| 115 |
+
"transformer.layers.10.self_attn.q_proj.biases": "model-00006-of-00024.safetensors",
|
| 116 |
+
"transformer.layers.10.self_attn.q_proj.scales": "model-00006-of-00024.safetensors",
|
| 117 |
+
"transformer.layers.10.self_attn.q_proj.weight": "model-00006-of-00024.safetensors",
|
| 118 |
+
"transformer.layers.10.self_attn.v_proj.biases": "model-00006-of-00024.safetensors",
|
| 119 |
+
"transformer.layers.10.self_attn.v_proj.scales": "model-00006-of-00024.safetensors",
|
| 120 |
+
"transformer.layers.10.self_attn.v_proj.weight": "model-00006-of-00024.safetensors",
|
| 121 |
+
"transformer.layers.11.input_layernorm.weight": "model-00006-of-00024.safetensors",
|
| 122 |
+
"transformer.layers.11.mlp.down_proj.biases": "model-00006-of-00024.safetensors",
|
| 123 |
+
"transformer.layers.11.mlp.down_proj.scales": "model-00006-of-00024.safetensors",
|
| 124 |
+
"transformer.layers.11.mlp.down_proj.weight": "model-00006-of-00024.safetensors",
|
| 125 |
+
"transformer.layers.11.mlp.gate_proj.biases": "model-00006-of-00024.safetensors",
|
| 126 |
+
"transformer.layers.11.mlp.gate_proj.scales": "model-00006-of-00024.safetensors",
|
| 127 |
+
"transformer.layers.11.mlp.gate_proj.weight": "model-00006-of-00024.safetensors",
|
| 128 |
+
"transformer.layers.11.mlp.up_proj.biases": "model-00006-of-00024.safetensors",
|
| 129 |
+
"transformer.layers.11.mlp.up_proj.scales": "model-00006-of-00024.safetensors",
|
| 130 |
+
"transformer.layers.11.mlp.up_proj.weight": "model-00007-of-00024.safetensors",
|
| 131 |
+
"transformer.layers.11.post_attention_layernorm.weight": "model-00007-of-00024.safetensors",
|
| 132 |
+
"transformer.layers.11.self_attn.k_proj.biases": "model-00007-of-00024.safetensors",
|
| 133 |
+
"transformer.layers.11.self_attn.k_proj.scales": "model-00007-of-00024.safetensors",
|
| 134 |
+
"transformer.layers.11.self_attn.k_proj.weight": "model-00007-of-00024.safetensors",
|
| 135 |
+
"transformer.layers.11.self_attn.o_proj.biases": "model-00007-of-00024.safetensors",
|
| 136 |
+
"transformer.layers.11.self_attn.o_proj.scales": "model-00007-of-00024.safetensors",
|
| 137 |
+
"transformer.layers.11.self_attn.o_proj.weight": "model-00007-of-00024.safetensors",
|
| 138 |
+
"transformer.layers.11.self_attn.q_proj.biases": "model-00007-of-00024.safetensors",
|
| 139 |
+
"transformer.layers.11.self_attn.q_proj.scales": "model-00007-of-00024.safetensors",
|
| 140 |
+
"transformer.layers.11.self_attn.q_proj.weight": "model-00007-of-00024.safetensors",
|
| 141 |
+
"transformer.layers.11.self_attn.v_proj.biases": "model-00007-of-00024.safetensors",
|
| 142 |
+
"transformer.layers.11.self_attn.v_proj.scales": "model-00007-of-00024.safetensors",
|
| 143 |
+
"transformer.layers.11.self_attn.v_proj.weight": "model-00007-of-00024.safetensors",
|
| 144 |
+
"transformer.layers.12.input_layernorm.weight": "model-00007-of-00024.safetensors",
|
| 145 |
+
"transformer.layers.12.mlp.down_proj.biases": "model-00007-of-00024.safetensors",
|
| 146 |
+
"transformer.layers.12.mlp.down_proj.scales": "model-00007-of-00024.safetensors",
|
| 147 |
+
"transformer.layers.12.mlp.down_proj.weight": "model-00007-of-00024.safetensors",
|
| 148 |
+
"transformer.layers.12.mlp.gate_proj.biases": "model-00007-of-00024.safetensors",
|
| 149 |
+
"transformer.layers.12.mlp.gate_proj.scales": "model-00007-of-00024.safetensors",
|
| 150 |
+
"transformer.layers.12.mlp.gate_proj.weight": "model-00007-of-00024.safetensors",
|
| 151 |
+
"transformer.layers.12.mlp.up_proj.biases": "model-00007-of-00024.safetensors",
|
| 152 |
+
"transformer.layers.12.mlp.up_proj.scales": "model-00007-of-00024.safetensors",
|
| 153 |
+
"transformer.layers.12.mlp.up_proj.weight": "model-00007-of-00024.safetensors",
|
| 154 |
+
"transformer.layers.12.post_attention_layernorm.weight": "model-00007-of-00024.safetensors",
|
| 155 |
+
"transformer.layers.12.self_attn.k_proj.biases": "model-00007-of-00024.safetensors",
|
| 156 |
+
"transformer.layers.12.self_attn.k_proj.scales": "model-00007-of-00024.safetensors",
|
| 157 |
+
"transformer.layers.12.self_attn.k_proj.weight": "model-00007-of-00024.safetensors",
|
| 158 |
+
"transformer.layers.12.self_attn.o_proj.biases": "model-00007-of-00024.safetensors",
|
| 159 |
+
"transformer.layers.12.self_attn.o_proj.scales": "model-00007-of-00024.safetensors",
|
| 160 |
+
"transformer.layers.12.self_attn.o_proj.weight": "model-00007-of-00024.safetensors",
|
| 161 |
+
"transformer.layers.12.self_attn.q_proj.biases": "model-00007-of-00024.safetensors",
|
| 162 |
+
"transformer.layers.12.self_attn.q_proj.scales": "model-00007-of-00024.safetensors",
|
| 163 |
+
"transformer.layers.12.self_attn.q_proj.weight": "model-00007-of-00024.safetensors",
|
| 164 |
+
"transformer.layers.12.self_attn.v_proj.biases": "model-00007-of-00024.safetensors",
|
| 165 |
+
"transformer.layers.12.self_attn.v_proj.scales": "model-00007-of-00024.safetensors",
|
| 166 |
+
"transformer.layers.12.self_attn.v_proj.weight": "model-00007-of-00024.safetensors",
|
| 167 |
+
"transformer.layers.13.input_layernorm.weight": "model-00007-of-00024.safetensors",
|
| 168 |
+
"transformer.layers.13.mlp.down_proj.biases": "model-00007-of-00024.safetensors",
|
| 169 |
+
"transformer.layers.13.mlp.down_proj.scales": "model-00007-of-00024.safetensors",
|
| 170 |
+
"transformer.layers.13.mlp.down_proj.weight": "model-00007-of-00024.safetensors",
|
| 171 |
+
"transformer.layers.13.mlp.gate_proj.biases": "model-00007-of-00024.safetensors",
|
| 172 |
+
"transformer.layers.13.mlp.gate_proj.scales": "model-00007-of-00024.safetensors",
|
| 173 |
+
"transformer.layers.13.mlp.gate_proj.weight": "model-00007-of-00024.safetensors",
|
| 174 |
+
"transformer.layers.13.mlp.up_proj.biases": "model-00007-of-00024.safetensors",
|
| 175 |
+
"transformer.layers.13.mlp.up_proj.scales": "model-00007-of-00024.safetensors",
|
| 176 |
+
"transformer.layers.13.mlp.up_proj.weight": "model-00007-of-00024.safetensors",
|
| 177 |
+
"transformer.layers.13.post_attention_layernorm.weight": "model-00007-of-00024.safetensors",
|
| 178 |
+
"transformer.layers.13.self_attn.k_proj.biases": "model-00007-of-00024.safetensors",
|
| 179 |
+
"transformer.layers.13.self_attn.k_proj.scales": "model-00007-of-00024.safetensors",
|
| 180 |
+
"transformer.layers.13.self_attn.k_proj.weight": "model-00007-of-00024.safetensors",
|
| 181 |
+
"transformer.layers.13.self_attn.o_proj.biases": "model-00007-of-00024.safetensors",
|
| 182 |
+
"transformer.layers.13.self_attn.o_proj.scales": "model-00007-of-00024.safetensors",
|
| 183 |
+
"transformer.layers.13.self_attn.o_proj.weight": "model-00007-of-00024.safetensors",
|
| 184 |
+
"transformer.layers.13.self_attn.q_proj.biases": "model-00007-of-00024.safetensors",
|
| 185 |
+
"transformer.layers.13.self_attn.q_proj.scales": "model-00007-of-00024.safetensors",
|
| 186 |
+
"transformer.layers.13.self_attn.q_proj.weight": "model-00008-of-00024.safetensors",
|
| 187 |
+
"transformer.layers.13.self_attn.v_proj.biases": "model-00008-of-00024.safetensors",
|
| 188 |
+
"transformer.layers.13.self_attn.v_proj.scales": "model-00008-of-00024.safetensors",
|
| 189 |
+
"transformer.layers.13.self_attn.v_proj.weight": "model-00008-of-00024.safetensors",
|
| 190 |
+
"transformer.layers.14.input_layernorm.weight": "model-00008-of-00024.safetensors",
|
| 191 |
+
"transformer.layers.14.mlp.down_proj.biases": "model-00008-of-00024.safetensors",
|
| 192 |
+
"transformer.layers.14.mlp.down_proj.scales": "model-00008-of-00024.safetensors",
|
| 193 |
+
"transformer.layers.14.mlp.down_proj.weight": "model-00008-of-00024.safetensors",
|
| 194 |
+
"transformer.layers.14.mlp.gate_proj.biases": "model-00008-of-00024.safetensors",
|
| 195 |
+
"transformer.layers.14.mlp.gate_proj.scales": "model-00008-of-00024.safetensors",
|
| 196 |
+
"transformer.layers.14.mlp.gate_proj.weight": "model-00008-of-00024.safetensors",
|
| 197 |
+
"transformer.layers.14.mlp.up_proj.biases": "model-00008-of-00024.safetensors",
|
| 198 |
+
"transformer.layers.14.mlp.up_proj.scales": "model-00008-of-00024.safetensors",
|
| 199 |
+
"transformer.layers.14.mlp.up_proj.weight": "model-00008-of-00024.safetensors",
|
| 200 |
+
"transformer.layers.14.post_attention_layernorm.weight": "model-00008-of-00024.safetensors",
|
| 201 |
+
"transformer.layers.14.self_attn.k_proj.biases": "model-00008-of-00024.safetensors",
|
| 202 |
+
"transformer.layers.14.self_attn.k_proj.scales": "model-00008-of-00024.safetensors",
|
| 203 |
+
"transformer.layers.14.self_attn.k_proj.weight": "model-00008-of-00024.safetensors",
|
| 204 |
+
"transformer.layers.14.self_attn.o_proj.biases": "model-00008-of-00024.safetensors",
|
| 205 |
+
"transformer.layers.14.self_attn.o_proj.scales": "model-00008-of-00024.safetensors",
|
| 206 |
+
"transformer.layers.14.self_attn.o_proj.weight": "model-00008-of-00024.safetensors",
|
| 207 |
+
"transformer.layers.14.self_attn.q_proj.biases": "model-00008-of-00024.safetensors",
|
| 208 |
+
"transformer.layers.14.self_attn.q_proj.scales": "model-00008-of-00024.safetensors",
|
| 209 |
+
"transformer.layers.14.self_attn.q_proj.weight": "model-00008-of-00024.safetensors",
|
| 210 |
+
"transformer.layers.14.self_attn.v_proj.biases": "model-00008-of-00024.safetensors",
|
| 211 |
+
"transformer.layers.14.self_attn.v_proj.scales": "model-00008-of-00024.safetensors",
|
| 212 |
+
"transformer.layers.14.self_attn.v_proj.weight": "model-00008-of-00024.safetensors",
|
| 213 |
+
"transformer.layers.15.input_layernorm.weight": "model-00008-of-00024.safetensors",
|
| 214 |
+
"transformer.layers.15.mlp.down_proj.biases": "model-00008-of-00024.safetensors",
|
| 215 |
+
"transformer.layers.15.mlp.down_proj.scales": "model-00008-of-00024.safetensors",
|
| 216 |
+
"transformer.layers.15.mlp.down_proj.weight": "model-00008-of-00024.safetensors",
|
| 217 |
+
"transformer.layers.15.mlp.gate_proj.biases": "model-00008-of-00024.safetensors",
|
| 218 |
+
"transformer.layers.15.mlp.gate_proj.scales": "model-00008-of-00024.safetensors",
|
| 219 |
+
"transformer.layers.15.mlp.gate_proj.weight": "model-00008-of-00024.safetensors",
|
| 220 |
+
"transformer.layers.15.mlp.up_proj.biases": "model-00008-of-00024.safetensors",
|
| 221 |
+
"transformer.layers.15.mlp.up_proj.scales": "model-00008-of-00024.safetensors",
|
| 222 |
+
"transformer.layers.15.mlp.up_proj.weight": "model-00008-of-00024.safetensors",
|
| 223 |
+
"transformer.layers.15.post_attention_layernorm.weight": "model-00008-of-00024.safetensors",
|
| 224 |
+
"transformer.layers.15.self_attn.k_proj.biases": "model-00008-of-00024.safetensors",
|
| 225 |
+
"transformer.layers.15.self_attn.k_proj.scales": "model-00008-of-00024.safetensors",
|
| 226 |
+
"transformer.layers.15.self_attn.k_proj.weight": "model-00008-of-00024.safetensors",
|
| 227 |
+
"transformer.layers.15.self_attn.o_proj.biases": "model-00008-of-00024.safetensors",
|
| 228 |
+
"transformer.layers.15.self_attn.o_proj.scales": "model-00008-of-00024.safetensors",
|
| 229 |
+
"transformer.layers.15.self_attn.o_proj.weight": "model-00008-of-00024.safetensors",
|
| 230 |
+
"transformer.layers.15.self_attn.q_proj.biases": "model-00008-of-00024.safetensors",
|
| 231 |
+
"transformer.layers.15.self_attn.q_proj.scales": "model-00008-of-00024.safetensors",
|
| 232 |
+
"transformer.layers.15.self_attn.q_proj.weight": "model-00008-of-00024.safetensors",
|
| 233 |
+
"transformer.layers.15.self_attn.v_proj.biases": "model-00008-of-00024.safetensors",
|
| 234 |
+
"transformer.layers.15.self_attn.v_proj.scales": "model-00008-of-00024.safetensors",
|
| 235 |
+
"transformer.layers.15.self_attn.v_proj.weight": "model-00008-of-00024.safetensors",
|
| 236 |
+
"transformer.layers.16.input_layernorm.weight": "model-00008-of-00024.safetensors",
|
| 237 |
+
"transformer.layers.16.mlp.down_proj.biases": "model-00008-of-00024.safetensors",
|
| 238 |
+
"transformer.layers.16.mlp.down_proj.scales": "model-00008-of-00024.safetensors",
|
| 239 |
+
"transformer.layers.16.mlp.down_proj.weight": "model-00009-of-00024.safetensors",
|
| 240 |
+
"transformer.layers.16.mlp.gate_proj.biases": "model-00009-of-00024.safetensors",
|
| 241 |
+
"transformer.layers.16.mlp.gate_proj.scales": "model-00009-of-00024.safetensors",
|
| 242 |
+
"transformer.layers.16.mlp.gate_proj.weight": "model-00009-of-00024.safetensors",
|
| 243 |
+
"transformer.layers.16.mlp.up_proj.biases": "model-00009-of-00024.safetensors",
|
| 244 |
+
"transformer.layers.16.mlp.up_proj.scales": "model-00009-of-00024.safetensors",
|
| 245 |
+
"transformer.layers.16.mlp.up_proj.weight": "model-00009-of-00024.safetensors",
|
| 246 |
+
"transformer.layers.16.post_attention_layernorm.weight": "model-00009-of-00024.safetensors",
|
| 247 |
+
"transformer.layers.16.self_attn.k_proj.biases": "model-00009-of-00024.safetensors",
|
| 248 |
+
"transformer.layers.16.self_attn.k_proj.scales": "model-00009-of-00024.safetensors",
|
| 249 |
+
"transformer.layers.16.self_attn.k_proj.weight": "model-00009-of-00024.safetensors",
|
| 250 |
+
"transformer.layers.16.self_attn.o_proj.biases": "model-00009-of-00024.safetensors",
|
| 251 |
+
"transformer.layers.16.self_attn.o_proj.scales": "model-00009-of-00024.safetensors",
|
| 252 |
+
"transformer.layers.16.self_attn.o_proj.weight": "model-00009-of-00024.safetensors",
|
| 253 |
+
"transformer.layers.16.self_attn.q_proj.biases": "model-00009-of-00024.safetensors",
|
| 254 |
+
"transformer.layers.16.self_attn.q_proj.scales": "model-00009-of-00024.safetensors",
|
| 255 |
+
"transformer.layers.16.self_attn.q_proj.weight": "model-00009-of-00024.safetensors",
|
| 256 |
+
"transformer.layers.16.self_attn.v_proj.biases": "model-00009-of-00024.safetensors",
|
| 257 |
+
"transformer.layers.16.self_attn.v_proj.scales": "model-00009-of-00024.safetensors",
|
| 258 |
+
"transformer.layers.16.self_attn.v_proj.weight": "model-00009-of-00024.safetensors",
|
| 259 |
+
"transformer.layers.17.input_layernorm.weight": "model-00009-of-00024.safetensors",
|
| 260 |
+
"transformer.layers.17.mlp.down_proj.biases": "model-00009-of-00024.safetensors",
|
| 261 |
+
"transformer.layers.17.mlp.down_proj.scales": "model-00009-of-00024.safetensors",
|
| 262 |
+
"transformer.layers.17.mlp.down_proj.weight": "model-00009-of-00024.safetensors",
|
| 263 |
+
"transformer.layers.17.mlp.gate_proj.biases": "model-00009-of-00024.safetensors",
|
| 264 |
+
"transformer.layers.17.mlp.gate_proj.scales": "model-00009-of-00024.safetensors",
|
| 265 |
+
"transformer.layers.17.mlp.gate_proj.weight": "model-00009-of-00024.safetensors",
|
| 266 |
+
"transformer.layers.17.mlp.up_proj.biases": "model-00009-of-00024.safetensors",
|
| 267 |
+
"transformer.layers.17.mlp.up_proj.scales": "model-00009-of-00024.safetensors",
|
| 268 |
+
"transformer.layers.17.mlp.up_proj.weight": "model-00009-of-00024.safetensors",
|
| 269 |
+
"transformer.layers.17.post_attention_layernorm.weight": "model-00009-of-00024.safetensors",
|
| 270 |
+
"transformer.layers.17.self_attn.k_proj.biases": "model-00009-of-00024.safetensors",
|
| 271 |
+
"transformer.layers.17.self_attn.k_proj.scales": "model-00009-of-00024.safetensors",
|
| 272 |
+
"transformer.layers.17.self_attn.k_proj.weight": "model-00009-of-00024.safetensors",
|
| 273 |
+
"transformer.layers.17.self_attn.o_proj.biases": "model-00009-of-00024.safetensors",
|
| 274 |
+
"transformer.layers.17.self_attn.o_proj.scales": "model-00009-of-00024.safetensors",
|
| 275 |
+
"transformer.layers.17.self_attn.o_proj.weight": "model-00009-of-00024.safetensors",
|
| 276 |
+
"transformer.layers.17.self_attn.q_proj.biases": "model-00009-of-00024.safetensors",
|
| 277 |
+
"transformer.layers.17.self_attn.q_proj.scales": "model-00009-of-00024.safetensors",
|
| 278 |
+
"transformer.layers.17.self_attn.q_proj.weight": "model-00009-of-00024.safetensors",
|
| 279 |
+
"transformer.layers.17.self_attn.v_proj.biases": "model-00009-of-00024.safetensors",
|
| 280 |
+
"transformer.layers.17.self_attn.v_proj.scales": "model-00009-of-00024.safetensors",
|
| 281 |
+
"transformer.layers.17.self_attn.v_proj.weight": "model-00009-of-00024.safetensors",
|
| 282 |
+
"transformer.layers.18.input_layernorm.weight": "model-00009-of-00024.safetensors",
|
| 283 |
+
"transformer.layers.18.mlp.down_proj.biases": "model-00009-of-00024.safetensors",
|
| 284 |
+
"transformer.layers.18.mlp.down_proj.scales": "model-00009-of-00024.safetensors",
|
| 285 |
+
"transformer.layers.18.mlp.down_proj.weight": "model-00009-of-00024.safetensors",
|
| 286 |
+
"transformer.layers.18.mlp.gate_proj.biases": "model-00009-of-00024.safetensors",
|
| 287 |
+
"transformer.layers.18.mlp.gate_proj.scales": "model-00009-of-00024.safetensors",
|
| 288 |
+
"transformer.layers.18.mlp.gate_proj.weight": "model-00010-of-00024.safetensors",
|
| 289 |
+
"transformer.layers.18.mlp.up_proj.biases": "model-00010-of-00024.safetensors",
|
| 290 |
+
"transformer.layers.18.mlp.up_proj.scales": "model-00010-of-00024.safetensors",
|
| 291 |
+
"transformer.layers.18.mlp.up_proj.weight": "model-00010-of-00024.safetensors",
|
| 292 |
+
"transformer.layers.18.post_attention_layernorm.weight": "model-00010-of-00024.safetensors",
|
| 293 |
+
"transformer.layers.18.self_attn.k_proj.biases": "model-00010-of-00024.safetensors",
|
| 294 |
+
"transformer.layers.18.self_attn.k_proj.scales": "model-00010-of-00024.safetensors",
|
| 295 |
+
"transformer.layers.18.self_attn.k_proj.weight": "model-00010-of-00024.safetensors",
|
| 296 |
+
"transformer.layers.18.self_attn.o_proj.biases": "model-00010-of-00024.safetensors",
|
| 297 |
+
"transformer.layers.18.self_attn.o_proj.scales": "model-00010-of-00024.safetensors",
|
| 298 |
+
"transformer.layers.18.self_attn.o_proj.weight": "model-00010-of-00024.safetensors",
|
| 299 |
+
"transformer.layers.18.self_attn.q_proj.biases": "model-00010-of-00024.safetensors",
|
| 300 |
+
"transformer.layers.18.self_attn.q_proj.scales": "model-00010-of-00024.safetensors",
|
| 301 |
+
"transformer.layers.18.self_attn.q_proj.weight": "model-00010-of-00024.safetensors",
|
| 302 |
+
"transformer.layers.18.self_attn.v_proj.biases": "model-00010-of-00024.safetensors",
|
| 303 |
+
"transformer.layers.18.self_attn.v_proj.scales": "model-00010-of-00024.safetensors",
|
| 304 |
+
"transformer.layers.18.self_attn.v_proj.weight": "model-00010-of-00024.safetensors",
|
| 305 |
+
"transformer.layers.19.input_layernorm.weight": "model-00010-of-00024.safetensors",
|
| 306 |
+
"transformer.layers.19.mlp.down_proj.biases": "model-00010-of-00024.safetensors",
|
| 307 |
+
"transformer.layers.19.mlp.down_proj.scales": "model-00010-of-00024.safetensors",
|
| 308 |
+
"transformer.layers.19.mlp.down_proj.weight": "model-00010-of-00024.safetensors",
|
| 309 |
+
"transformer.layers.19.mlp.gate_proj.biases": "model-00010-of-00024.safetensors",
|
| 310 |
+
"transformer.layers.19.mlp.gate_proj.scales": "model-00010-of-00024.safetensors",
|
| 311 |
+
"transformer.layers.19.mlp.gate_proj.weight": "model-00010-of-00024.safetensors",
|
| 312 |
+
"transformer.layers.19.mlp.up_proj.biases": "model-00010-of-00024.safetensors",
|
| 313 |
+
"transformer.layers.19.mlp.up_proj.scales": "model-00010-of-00024.safetensors",
|
| 314 |
+
"transformer.layers.19.mlp.up_proj.weight": "model-00010-of-00024.safetensors",
|
| 315 |
+
"transformer.layers.19.post_attention_layernorm.weight": "model-00010-of-00024.safetensors",
|
| 316 |
+
"transformer.layers.19.self_attn.k_proj.biases": "model-00010-of-00024.safetensors",
|
| 317 |
+
"transformer.layers.19.self_attn.k_proj.scales": "model-00010-of-00024.safetensors",
|
| 318 |
+
"transformer.layers.19.self_attn.k_proj.weight": "model-00010-of-00024.safetensors",
|
| 319 |
+
"transformer.layers.19.self_attn.o_proj.biases": "model-00010-of-00024.safetensors",
|
| 320 |
+
"transformer.layers.19.self_attn.o_proj.scales": "model-00010-of-00024.safetensors",
|
| 321 |
+
"transformer.layers.19.self_attn.o_proj.weight": "model-00010-of-00024.safetensors",
|
| 322 |
+
"transformer.layers.19.self_attn.q_proj.biases": "model-00010-of-00024.safetensors",
|
| 323 |
+
"transformer.layers.19.self_attn.q_proj.scales": "model-00010-of-00024.safetensors",
|
| 324 |
+
"transformer.layers.19.self_attn.q_proj.weight": "model-00010-of-00024.safetensors",
|
| 325 |
+
"transformer.layers.19.self_attn.v_proj.biases": "model-00010-of-00024.safetensors",
|
| 326 |
+
"transformer.layers.19.self_attn.v_proj.scales": "model-00010-of-00024.safetensors",
|
| 327 |
+
"transformer.layers.19.self_attn.v_proj.weight": "model-00010-of-00024.safetensors",
|
| 328 |
+
"transformer.layers.2.input_layernorm.weight": "model-00010-of-00024.safetensors",
|
| 329 |
+
"transformer.layers.2.mlp.down_proj.biases": "model-00010-of-00024.safetensors",
|
| 330 |
+
"transformer.layers.2.mlp.down_proj.scales": "model-00010-of-00024.safetensors",
|
| 331 |
+
"transformer.layers.2.mlp.down_proj.weight": "model-00010-of-00024.safetensors",
|
| 332 |
+
"transformer.layers.2.mlp.gate_proj.biases": "model-00010-of-00024.safetensors",
|
| 333 |
+
"transformer.layers.2.mlp.gate_proj.scales": "model-00010-of-00024.safetensors",
|
| 334 |
+
"transformer.layers.2.mlp.gate_proj.weight": "model-00010-of-00024.safetensors",
|
| 335 |
+
"transformer.layers.2.mlp.up_proj.biases": "model-00010-of-00024.safetensors",
|
| 336 |
+
"transformer.layers.2.mlp.up_proj.scales": "model-00010-of-00024.safetensors",
|
| 337 |
+
"transformer.layers.2.mlp.up_proj.weight": "model-00011-of-00024.safetensors",
|
| 338 |
+
"transformer.layers.2.post_attention_layernorm.weight": "model-00011-of-00024.safetensors",
|
| 339 |
+
"transformer.layers.2.self_attn.k_proj.biases": "model-00011-of-00024.safetensors",
|
| 340 |
+
"transformer.layers.2.self_attn.k_proj.scales": "model-00011-of-00024.safetensors",
|
| 341 |
+
"transformer.layers.2.self_attn.k_proj.weight": "model-00011-of-00024.safetensors",
|
| 342 |
+
"transformer.layers.2.self_attn.o_proj.biases": "model-00011-of-00024.safetensors",
|
| 343 |
+
"transformer.layers.2.self_attn.o_proj.scales": "model-00011-of-00024.safetensors",
|
| 344 |
+
"transformer.layers.2.self_attn.o_proj.weight": "model-00011-of-00024.safetensors",
|
| 345 |
+
"transformer.layers.2.self_attn.q_proj.biases": "model-00011-of-00024.safetensors",
|
| 346 |
+
"transformer.layers.2.self_attn.q_proj.scales": "model-00011-of-00024.safetensors",
|
| 347 |
+
"transformer.layers.2.self_attn.q_proj.weight": "model-00011-of-00024.safetensors",
|
| 348 |
+
"transformer.layers.2.self_attn.v_proj.biases": "model-00011-of-00024.safetensors",
|
| 349 |
+
"transformer.layers.2.self_attn.v_proj.scales": "model-00011-of-00024.safetensors",
|
| 350 |
+
"transformer.layers.2.self_attn.v_proj.weight": "model-00011-of-00024.safetensors",
|
| 351 |
+
"transformer.layers.20.input_layernorm.weight": "model-00011-of-00024.safetensors",
|
| 352 |
+
"transformer.layers.20.mlp.down_proj.biases": "model-00011-of-00024.safetensors",
|
| 353 |
+
"transformer.layers.20.mlp.down_proj.scales": "model-00011-of-00024.safetensors",
|
| 354 |
+
"transformer.layers.20.mlp.down_proj.weight": "model-00011-of-00024.safetensors",
|
| 355 |
+
"transformer.layers.20.mlp.gate_proj.biases": "model-00011-of-00024.safetensors",
|
| 356 |
+
"transformer.layers.20.mlp.gate_proj.scales": "model-00011-of-00024.safetensors",
|
| 357 |
+
"transformer.layers.20.mlp.gate_proj.weight": "model-00011-of-00024.safetensors",
|
| 358 |
+
"transformer.layers.20.mlp.up_proj.biases": "model-00011-of-00024.safetensors",
|
| 359 |
+
"transformer.layers.20.mlp.up_proj.scales": "model-00011-of-00024.safetensors",
|
| 360 |
+
"transformer.layers.20.mlp.up_proj.weight": "model-00011-of-00024.safetensors",
|
| 361 |
+
"transformer.layers.20.post_attention_layernorm.weight": "model-00011-of-00024.safetensors",
|
| 362 |
+
"transformer.layers.20.self_attn.k_proj.biases": "model-00011-of-00024.safetensors",
|
| 363 |
+
"transformer.layers.20.self_attn.k_proj.scales": "model-00011-of-00024.safetensors",
|
| 364 |
+
"transformer.layers.20.self_attn.k_proj.weight": "model-00011-of-00024.safetensors",
|
| 365 |
+
"transformer.layers.20.self_attn.o_proj.biases": "model-00011-of-00024.safetensors",
|
| 366 |
+
"transformer.layers.20.self_attn.o_proj.scales": "model-00011-of-00024.safetensors",
|
| 367 |
+
"transformer.layers.20.self_attn.o_proj.weight": "model-00011-of-00024.safetensors",
|
| 368 |
+
"transformer.layers.20.self_attn.q_proj.biases": "model-00011-of-00024.safetensors",
|
| 369 |
+
"transformer.layers.20.self_attn.q_proj.scales": "model-00011-of-00024.safetensors",
|
| 370 |
+
"transformer.layers.20.self_attn.q_proj.weight": "model-00011-of-00024.safetensors",
|
| 371 |
+
"transformer.layers.20.self_attn.v_proj.biases": "model-00011-of-00024.safetensors",
|
| 372 |
+
"transformer.layers.20.self_attn.v_proj.scales": "model-00011-of-00024.safetensors",
|
| 373 |
+
"transformer.layers.20.self_attn.v_proj.weight": "model-00011-of-00024.safetensors",
|
| 374 |
+
"transformer.layers.21.input_layernorm.weight": "model-00011-of-00024.safetensors",
|
| 375 |
+
"transformer.layers.21.mlp.down_proj.biases": "model-00011-of-00024.safetensors",
|
| 376 |
+
"transformer.layers.21.mlp.down_proj.scales": "model-00011-of-00024.safetensors",
|
| 377 |
+
"transformer.layers.21.mlp.down_proj.weight": "model-00011-of-00024.safetensors",
|
| 378 |
+
"transformer.layers.21.mlp.gate_proj.biases": "model-00011-of-00024.safetensors",
|
| 379 |
+
"transformer.layers.21.mlp.gate_proj.scales": "model-00011-of-00024.safetensors",
|
| 380 |
+
"transformer.layers.21.mlp.gate_proj.weight": "model-00011-of-00024.safetensors",
|
| 381 |
+
"transformer.layers.21.mlp.up_proj.biases": "model-00011-of-00024.safetensors",
|
| 382 |
+
"transformer.layers.21.mlp.up_proj.scales": "model-00011-of-00024.safetensors",
|
| 383 |
+
"transformer.layers.21.mlp.up_proj.weight": "model-00011-of-00024.safetensors",
|
| 384 |
+
"transformer.layers.21.post_attention_layernorm.weight": "model-00011-of-00024.safetensors",
|
| 385 |
+
"transformer.layers.21.self_attn.k_proj.biases": "model-00011-of-00024.safetensors",
|
| 386 |
+
"transformer.layers.21.self_attn.k_proj.scales": "model-00011-of-00024.safetensors",
|
| 387 |
+
"transformer.layers.21.self_attn.k_proj.weight": "model-00011-of-00024.safetensors",
|
| 388 |
+
"transformer.layers.21.self_attn.o_proj.biases": "model-00011-of-00024.safetensors",
|
| 389 |
+
"transformer.layers.21.self_attn.o_proj.scales": "model-00011-of-00024.safetensors",
|
| 390 |
+
"transformer.layers.21.self_attn.o_proj.weight": "model-00011-of-00024.safetensors",
|
| 391 |
+
"transformer.layers.21.self_attn.q_proj.biases": "model-00011-of-00024.safetensors",
|
| 392 |
+
"transformer.layers.21.self_attn.q_proj.scales": "model-00011-of-00024.safetensors",
|
| 393 |
+
"transformer.layers.21.self_attn.q_proj.weight": "model-00012-of-00024.safetensors",
|
| 394 |
+
"transformer.layers.21.self_attn.v_proj.biases": "model-00012-of-00024.safetensors",
|
| 395 |
+
"transformer.layers.21.self_attn.v_proj.scales": "model-00012-of-00024.safetensors",
|
| 396 |
+
"transformer.layers.21.self_attn.v_proj.weight": "model-00012-of-00024.safetensors",
|
| 397 |
+
"transformer.layers.22.input_layernorm.weight": "model-00012-of-00024.safetensors",
|
| 398 |
+
"transformer.layers.22.mlp.down_proj.biases": "model-00012-of-00024.safetensors",
|
| 399 |
+
"transformer.layers.22.mlp.down_proj.scales": "model-00012-of-00024.safetensors",
|
| 400 |
+
"transformer.layers.22.mlp.down_proj.weight": "model-00012-of-00024.safetensors",
|
| 401 |
+
"transformer.layers.22.mlp.gate_proj.biases": "model-00012-of-00024.safetensors",
|
| 402 |
+
"transformer.layers.22.mlp.gate_proj.scales": "model-00012-of-00024.safetensors",
|
| 403 |
+
"transformer.layers.22.mlp.gate_proj.weight": "model-00012-of-00024.safetensors",
|
| 404 |
+
"transformer.layers.22.mlp.up_proj.biases": "model-00012-of-00024.safetensors",
|
| 405 |
+
"transformer.layers.22.mlp.up_proj.scales": "model-00012-of-00024.safetensors",
|
| 406 |
+
"transformer.layers.22.mlp.up_proj.weight": "model-00012-of-00024.safetensors",
|
| 407 |
+
"transformer.layers.22.post_attention_layernorm.weight": "model-00012-of-00024.safetensors",
|
| 408 |
+
"transformer.layers.22.self_attn.k_proj.biases": "model-00012-of-00024.safetensors",
|
| 409 |
+
"transformer.layers.22.self_attn.k_proj.scales": "model-00012-of-00024.safetensors",
|
| 410 |
+
"transformer.layers.22.self_attn.k_proj.weight": "model-00012-of-00024.safetensors",
|
| 411 |
+
"transformer.layers.22.self_attn.o_proj.biases": "model-00012-of-00024.safetensors",
|
| 412 |
+
"transformer.layers.22.self_attn.o_proj.scales": "model-00012-of-00024.safetensors",
|
| 413 |
+
"transformer.layers.22.self_attn.o_proj.weight": "model-00012-of-00024.safetensors",
|
| 414 |
+
"transformer.layers.22.self_attn.q_proj.biases": "model-00012-of-00024.safetensors",
|
| 415 |
+
"transformer.layers.22.self_attn.q_proj.scales": "model-00012-of-00024.safetensors",
|
| 416 |
+
"transformer.layers.22.self_attn.q_proj.weight": "model-00012-of-00024.safetensors",
|
| 417 |
+
"transformer.layers.22.self_attn.v_proj.biases": "model-00012-of-00024.safetensors",
|
| 418 |
+
"transformer.layers.22.self_attn.v_proj.scales": "model-00012-of-00024.safetensors",
|
| 419 |
+
"transformer.layers.22.self_attn.v_proj.weight": "model-00012-of-00024.safetensors",
|
| 420 |
+
"transformer.layers.23.input_layernorm.weight": "model-00012-of-00024.safetensors",
|
| 421 |
+
"transformer.layers.23.mlp.down_proj.biases": "model-00012-of-00024.safetensors",
|
| 422 |
+
"transformer.layers.23.mlp.down_proj.scales": "model-00012-of-00024.safetensors",
|
| 423 |
+
"transformer.layers.23.mlp.down_proj.weight": "model-00012-of-00024.safetensors",
|
| 424 |
+
"transformer.layers.23.mlp.gate_proj.biases": "model-00012-of-00024.safetensors",
|
| 425 |
+
"transformer.layers.23.mlp.gate_proj.scales": "model-00012-of-00024.safetensors",
|
| 426 |
+
"transformer.layers.23.mlp.gate_proj.weight": "model-00012-of-00024.safetensors",
|
| 427 |
+
"transformer.layers.23.mlp.up_proj.biases": "model-00012-of-00024.safetensors",
|
| 428 |
+
"transformer.layers.23.mlp.up_proj.scales": "model-00012-of-00024.safetensors",
|
| 429 |
+
"transformer.layers.23.mlp.up_proj.weight": "model-00012-of-00024.safetensors",
|
| 430 |
+
"transformer.layers.23.post_attention_layernorm.weight": "model-00012-of-00024.safetensors",
|
| 431 |
+
"transformer.layers.23.self_attn.k_proj.biases": "model-00012-of-00024.safetensors",
|
| 432 |
+
"transformer.layers.23.self_attn.k_proj.scales": "model-00012-of-00024.safetensors",
|
| 433 |
+
"transformer.layers.23.self_attn.k_proj.weight": "model-00012-of-00024.safetensors",
|
| 434 |
+
"transformer.layers.23.self_attn.o_proj.biases": "model-00012-of-00024.safetensors",
|
| 435 |
+
"transformer.layers.23.self_attn.o_proj.scales": "model-00012-of-00024.safetensors",
|
| 436 |
+
"transformer.layers.23.self_attn.o_proj.weight": "model-00012-of-00024.safetensors",
|
| 437 |
+
"transformer.layers.23.self_attn.q_proj.biases": "model-00012-of-00024.safetensors",
|
| 438 |
+
"transformer.layers.23.self_attn.q_proj.scales": "model-00012-of-00024.safetensors",
|
| 439 |
+
"transformer.layers.23.self_attn.q_proj.weight": "model-00012-of-00024.safetensors",
|
| 440 |
+
"transformer.layers.23.self_attn.v_proj.biases": "model-00012-of-00024.safetensors",
|
| 441 |
+
"transformer.layers.23.self_attn.v_proj.scales": "model-00012-of-00024.safetensors",
|
| 442 |
+
"transformer.layers.23.self_attn.v_proj.weight": "model-00012-of-00024.safetensors",
|
| 443 |
+
"transformer.layers.24.input_layernorm.weight": "model-00012-of-00024.safetensors",
|
| 444 |
+
"transformer.layers.24.mlp.down_proj.biases": "model-00012-of-00024.safetensors",
|
| 445 |
+
"transformer.layers.24.mlp.down_proj.scales": "model-00012-of-00024.safetensors",
|
| 446 |
+
"transformer.layers.24.mlp.down_proj.weight": "model-00013-of-00024.safetensors",
|
| 447 |
+
"transformer.layers.24.mlp.gate_proj.biases": "model-00013-of-00024.safetensors",
|
| 448 |
+
"transformer.layers.24.mlp.gate_proj.scales": "model-00013-of-00024.safetensors",
|
| 449 |
+
"transformer.layers.24.mlp.gate_proj.weight": "model-00013-of-00024.safetensors",
|
| 450 |
+
"transformer.layers.24.mlp.up_proj.biases": "model-00013-of-00024.safetensors",
|
| 451 |
+
"transformer.layers.24.mlp.up_proj.scales": "model-00013-of-00024.safetensors",
|
| 452 |
+
"transformer.layers.24.mlp.up_proj.weight": "model-00013-of-00024.safetensors",
|
| 453 |
+
"transformer.layers.24.post_attention_layernorm.weight": "model-00013-of-00024.safetensors",
|
| 454 |
+
"transformer.layers.24.self_attn.k_proj.biases": "model-00013-of-00024.safetensors",
|
| 455 |
+
"transformer.layers.24.self_attn.k_proj.scales": "model-00013-of-00024.safetensors",
|
| 456 |
+
"transformer.layers.24.self_attn.k_proj.weight": "model-00013-of-00024.safetensors",
|
| 457 |
+
"transformer.layers.24.self_attn.o_proj.biases": "model-00013-of-00024.safetensors",
|
| 458 |
+
"transformer.layers.24.self_attn.o_proj.scales": "model-00013-of-00024.safetensors",
|
| 459 |
+
"transformer.layers.24.self_attn.o_proj.weight": "model-00013-of-00024.safetensors",
|
| 460 |
+
"transformer.layers.24.self_attn.q_proj.biases": "model-00013-of-00024.safetensors",
|
| 461 |
+
"transformer.layers.24.self_attn.q_proj.scales": "model-00013-of-00024.safetensors",
|
| 462 |
+
"transformer.layers.24.self_attn.q_proj.weight": "model-00013-of-00024.safetensors",
|
| 463 |
+
"transformer.layers.24.self_attn.v_proj.biases": "model-00013-of-00024.safetensors",
|
| 464 |
+
"transformer.layers.24.self_attn.v_proj.scales": "model-00013-of-00024.safetensors",
|
| 465 |
+
"transformer.layers.24.self_attn.v_proj.weight": "model-00013-of-00024.safetensors",
|
| 466 |
+
"transformer.layers.25.input_layernorm.weight": "model-00013-of-00024.safetensors",
|
| 467 |
+
"transformer.layers.25.mlp.down_proj.biases": "model-00013-of-00024.safetensors",
|
| 468 |
+
"transformer.layers.25.mlp.down_proj.scales": "model-00013-of-00024.safetensors",
|
| 469 |
+
"transformer.layers.25.mlp.down_proj.weight": "model-00013-of-00024.safetensors",
|
| 470 |
+
"transformer.layers.25.mlp.gate_proj.biases": "model-00013-of-00024.safetensors",
|
| 471 |
+
"transformer.layers.25.mlp.gate_proj.scales": "model-00013-of-00024.safetensors",
|
| 472 |
+
"transformer.layers.25.mlp.gate_proj.weight": "model-00013-of-00024.safetensors",
|
| 473 |
+
"transformer.layers.25.mlp.up_proj.biases": "model-00013-of-00024.safetensors",
|
| 474 |
+
"transformer.layers.25.mlp.up_proj.scales": "model-00013-of-00024.safetensors",
|
| 475 |
+
"transformer.layers.25.mlp.up_proj.weight": "model-00013-of-00024.safetensors",
|
| 476 |
+
"transformer.layers.25.post_attention_layernorm.weight": "model-00013-of-00024.safetensors",
|
| 477 |
+
"transformer.layers.25.self_attn.k_proj.biases": "model-00013-of-00024.safetensors",
|
| 478 |
+
"transformer.layers.25.self_attn.k_proj.scales": "model-00013-of-00024.safetensors",
|
| 479 |
+
"transformer.layers.25.self_attn.k_proj.weight": "model-00013-of-00024.safetensors",
|
| 480 |
+
"transformer.layers.25.self_attn.o_proj.biases": "model-00013-of-00024.safetensors",
|
| 481 |
+
"transformer.layers.25.self_attn.o_proj.scales": "model-00013-of-00024.safetensors",
|
| 482 |
+
"transformer.layers.25.self_attn.o_proj.weight": "model-00013-of-00024.safetensors",
|
| 483 |
+
"transformer.layers.25.self_attn.q_proj.biases": "model-00013-of-00024.safetensors",
|
| 484 |
+
"transformer.layers.25.self_attn.q_proj.scales": "model-00013-of-00024.safetensors",
|
| 485 |
+
"transformer.layers.25.self_attn.q_proj.weight": "model-00013-of-00024.safetensors",
|
| 486 |
+
"transformer.layers.25.self_attn.v_proj.biases": "model-00013-of-00024.safetensors",
|
| 487 |
+
"transformer.layers.25.self_attn.v_proj.scales": "model-00013-of-00024.safetensors",
|
| 488 |
+
"transformer.layers.25.self_attn.v_proj.weight": "model-00013-of-00024.safetensors",
|
| 489 |
+
"transformer.layers.26.input_layernorm.weight": "model-00013-of-00024.safetensors",
|
| 490 |
+
"transformer.layers.26.mlp.down_proj.biases": "model-00013-of-00024.safetensors",
|
| 491 |
+
"transformer.layers.26.mlp.down_proj.scales": "model-00013-of-00024.safetensors",
|
| 492 |
+
"transformer.layers.26.mlp.down_proj.weight": "model-00013-of-00024.safetensors",
|
| 493 |
+
"transformer.layers.26.mlp.gate_proj.biases": "model-00013-of-00024.safetensors",
|
| 494 |
+
"transformer.layers.26.mlp.gate_proj.scales": "model-00013-of-00024.safetensors",
|
| 495 |
+
"transformer.layers.26.mlp.gate_proj.weight": "model-00014-of-00024.safetensors",
|
| 496 |
+
"transformer.layers.26.mlp.up_proj.biases": "model-00014-of-00024.safetensors",
|
| 497 |
+
"transformer.layers.26.mlp.up_proj.scales": "model-00014-of-00024.safetensors",
|
| 498 |
+
"transformer.layers.26.mlp.up_proj.weight": "model-00014-of-00024.safetensors",
|
| 499 |
+
"transformer.layers.26.post_attention_layernorm.weight": "model-00014-of-00024.safetensors",
|
| 500 |
+
"transformer.layers.26.self_attn.k_proj.biases": "model-00014-of-00024.safetensors",
|
| 501 |
+
"transformer.layers.26.self_attn.k_proj.scales": "model-00014-of-00024.safetensors",
|
| 502 |
+
"transformer.layers.26.self_attn.k_proj.weight": "model-00014-of-00024.safetensors",
|
| 503 |
+
"transformer.layers.26.self_attn.o_proj.biases": "model-00014-of-00024.safetensors",
|
| 504 |
+
"transformer.layers.26.self_attn.o_proj.scales": "model-00014-of-00024.safetensors",
|
| 505 |
+
"transformer.layers.26.self_attn.o_proj.weight": "model-00014-of-00024.safetensors",
|
| 506 |
+
"transformer.layers.26.self_attn.q_proj.biases": "model-00014-of-00024.safetensors",
|
| 507 |
+
"transformer.layers.26.self_attn.q_proj.scales": "model-00014-of-00024.safetensors",
|
| 508 |
+
"transformer.layers.26.self_attn.q_proj.weight": "model-00014-of-00024.safetensors",
|
| 509 |
+
"transformer.layers.26.self_attn.v_proj.biases": "model-00014-of-00024.safetensors",
|
| 510 |
+
"transformer.layers.26.self_attn.v_proj.scales": "model-00014-of-00024.safetensors",
|
| 511 |
+
"transformer.layers.26.self_attn.v_proj.weight": "model-00014-of-00024.safetensors",
|
| 512 |
+
"transformer.layers.27.input_layernorm.weight": "model-00014-of-00024.safetensors",
|
| 513 |
+
"transformer.layers.27.mlp.down_proj.biases": "model-00014-of-00024.safetensors",
|
| 514 |
+
"transformer.layers.27.mlp.down_proj.scales": "model-00014-of-00024.safetensors",
|
| 515 |
+
"transformer.layers.27.mlp.down_proj.weight": "model-00014-of-00024.safetensors",
|
| 516 |
+
"transformer.layers.27.mlp.gate_proj.biases": "model-00014-of-00024.safetensors",
|
| 517 |
+
"transformer.layers.27.mlp.gate_proj.scales": "model-00014-of-00024.safetensors",
|
| 518 |
+
"transformer.layers.27.mlp.gate_proj.weight": "model-00014-of-00024.safetensors",
|
| 519 |
+
"transformer.layers.27.mlp.up_proj.biases": "model-00014-of-00024.safetensors",
|
| 520 |
+
"transformer.layers.27.mlp.up_proj.scales": "model-00014-of-00024.safetensors",
|
| 521 |
+
"transformer.layers.27.mlp.up_proj.weight": "model-00014-of-00024.safetensors",
|
| 522 |
+
"transformer.layers.27.post_attention_layernorm.weight": "model-00014-of-00024.safetensors",
|
| 523 |
+
"transformer.layers.27.self_attn.k_proj.biases": "model-00014-of-00024.safetensors",
|
| 524 |
+
"transformer.layers.27.self_attn.k_proj.scales": "model-00014-of-00024.safetensors",
|
| 525 |
+
"transformer.layers.27.self_attn.k_proj.weight": "model-00014-of-00024.safetensors",
|
| 526 |
+
"transformer.layers.27.self_attn.o_proj.biases": "model-00014-of-00024.safetensors",
|
| 527 |
+
"transformer.layers.27.self_attn.o_proj.scales": "model-00014-of-00024.safetensors",
|
| 528 |
+
"transformer.layers.27.self_attn.o_proj.weight": "model-00014-of-00024.safetensors",
|
| 529 |
+
"transformer.layers.27.self_attn.q_proj.biases": "model-00014-of-00024.safetensors",
|
| 530 |
+
"transformer.layers.27.self_attn.q_proj.scales": "model-00014-of-00024.safetensors",
|
| 531 |
+
"transformer.layers.27.self_attn.q_proj.weight": "model-00014-of-00024.safetensors",
|
| 532 |
+
"transformer.layers.27.self_attn.v_proj.biases": "model-00014-of-00024.safetensors",
|
| 533 |
+
"transformer.layers.27.self_attn.v_proj.scales": "model-00014-of-00024.safetensors",
|
| 534 |
+
"transformer.layers.27.self_attn.v_proj.weight": "model-00014-of-00024.safetensors",
|
| 535 |
+
"transformer.layers.3.input_layernorm.weight": "model-00014-of-00024.safetensors",
|
| 536 |
+
"transformer.layers.3.mlp.down_proj.biases": "model-00014-of-00024.safetensors",
|
| 537 |
+
"transformer.layers.3.mlp.down_proj.scales": "model-00014-of-00024.safetensors",
|
| 538 |
+
"transformer.layers.3.mlp.down_proj.weight": "model-00014-of-00024.safetensors",
|
| 539 |
+
"transformer.layers.3.mlp.gate_proj.biases": "model-00014-of-00024.safetensors",
|
| 540 |
+
"transformer.layers.3.mlp.gate_proj.scales": "model-00014-of-00024.safetensors",
|
| 541 |
+
"transformer.layers.3.mlp.gate_proj.weight": "model-00014-of-00024.safetensors",
|
| 542 |
+
"transformer.layers.3.mlp.up_proj.biases": "model-00014-of-00024.safetensors",
|
| 543 |
+
"transformer.layers.3.mlp.up_proj.scales": "model-00014-of-00024.safetensors",
|
| 544 |
+
"transformer.layers.3.mlp.up_proj.weight": "model-00015-of-00024.safetensors",
|
| 545 |
+
"transformer.layers.3.post_attention_layernorm.weight": "model-00015-of-00024.safetensors",
|
| 546 |
+
"transformer.layers.3.self_attn.k_proj.biases": "model-00015-of-00024.safetensors",
|
| 547 |
+
"transformer.layers.3.self_attn.k_proj.scales": "model-00015-of-00024.safetensors",
|
| 548 |
+
"transformer.layers.3.self_attn.k_proj.weight": "model-00015-of-00024.safetensors",
|
| 549 |
+
"transformer.layers.3.self_attn.o_proj.biases": "model-00015-of-00024.safetensors",
|
| 550 |
+
"transformer.layers.3.self_attn.o_proj.scales": "model-00015-of-00024.safetensors",
|
| 551 |
+
"transformer.layers.3.self_attn.o_proj.weight": "model-00015-of-00024.safetensors",
|
| 552 |
+
"transformer.layers.3.self_attn.q_proj.biases": "model-00015-of-00024.safetensors",
|
| 553 |
+
"transformer.layers.3.self_attn.q_proj.scales": "model-00015-of-00024.safetensors",
|
| 554 |
+
"transformer.layers.3.self_attn.q_proj.weight": "model-00015-of-00024.safetensors",
|
| 555 |
+
"transformer.layers.3.self_attn.v_proj.biases": "model-00015-of-00024.safetensors",
|
| 556 |
+
"transformer.layers.3.self_attn.v_proj.scales": "model-00015-of-00024.safetensors",
|
| 557 |
+
"transformer.layers.3.self_attn.v_proj.weight": "model-00015-of-00024.safetensors",
|
| 558 |
+
"transformer.layers.4.input_layernorm.weight": "model-00015-of-00024.safetensors",
|
| 559 |
+
"transformer.layers.4.mlp.down_proj.biases": "model-00015-of-00024.safetensors",
|
| 560 |
+
"transformer.layers.4.mlp.down_proj.scales": "model-00015-of-00024.safetensors",
|
| 561 |
+
"transformer.layers.4.mlp.down_proj.weight": "model-00015-of-00024.safetensors",
|
| 562 |
+
"transformer.layers.4.mlp.gate_proj.biases": "model-00015-of-00024.safetensors",
|
| 563 |
+
"transformer.layers.4.mlp.gate_proj.scales": "model-00015-of-00024.safetensors",
|
| 564 |
+
"transformer.layers.4.mlp.gate_proj.weight": "model-00015-of-00024.safetensors",
|
| 565 |
+
"transformer.layers.4.mlp.up_proj.biases": "model-00015-of-00024.safetensors",
|
| 566 |
+
"transformer.layers.4.mlp.up_proj.scales": "model-00015-of-00024.safetensors",
|
| 567 |
+
"transformer.layers.4.mlp.up_proj.weight": "model-00015-of-00024.safetensors",
|
| 568 |
+
"transformer.layers.4.post_attention_layernorm.weight": "model-00015-of-00024.safetensors",
|
| 569 |
+
"transformer.layers.4.self_attn.k_proj.biases": "model-00015-of-00024.safetensors",
|
| 570 |
+
"transformer.layers.4.self_attn.k_proj.scales": "model-00015-of-00024.safetensors",
|
| 571 |
+
"transformer.layers.4.self_attn.k_proj.weight": "model-00015-of-00024.safetensors",
|
| 572 |
+
"transformer.layers.4.self_attn.o_proj.biases": "model-00015-of-00024.safetensors",
|
| 573 |
+
"transformer.layers.4.self_attn.o_proj.scales": "model-00015-of-00024.safetensors",
|
| 574 |
+
"transformer.layers.4.self_attn.o_proj.weight": "model-00015-of-00024.safetensors",
|
| 575 |
+
"transformer.layers.4.self_attn.q_proj.biases": "model-00015-of-00024.safetensors",
|
| 576 |
+
"transformer.layers.4.self_attn.q_proj.scales": "model-00015-of-00024.safetensors",
|
| 577 |
+
"transformer.layers.4.self_attn.q_proj.weight": "model-00015-of-00024.safetensors",
|
| 578 |
+
"transformer.layers.4.self_attn.v_proj.biases": "model-00015-of-00024.safetensors",
|
| 579 |
+
"transformer.layers.4.self_attn.v_proj.scales": "model-00015-of-00024.safetensors",
|
| 580 |
+
"transformer.layers.4.self_attn.v_proj.weight": "model-00015-of-00024.safetensors",
|
| 581 |
+
"transformer.layers.5.input_layernorm.weight": "model-00015-of-00024.safetensors",
|
| 582 |
+
"transformer.layers.5.mlp.down_proj.biases": "model-00015-of-00024.safetensors",
|
| 583 |
+
"transformer.layers.5.mlp.down_proj.scales": "model-00015-of-00024.safetensors",
|
| 584 |
+
"transformer.layers.5.mlp.down_proj.weight": "model-00015-of-00024.safetensors",
|
| 585 |
+
"transformer.layers.5.mlp.gate_proj.biases": "model-00015-of-00024.safetensors",
|
| 586 |
+
"transformer.layers.5.mlp.gate_proj.scales": "model-00015-of-00024.safetensors",
|
| 587 |
+
"transformer.layers.5.mlp.gate_proj.weight": "model-00015-of-00024.safetensors",
|
| 588 |
+
"transformer.layers.5.mlp.up_proj.biases": "model-00015-of-00024.safetensors",
|
| 589 |
+
"transformer.layers.5.mlp.up_proj.scales": "model-00015-of-00024.safetensors",
|
| 590 |
+
"transformer.layers.5.mlp.up_proj.weight": "model-00015-of-00024.safetensors",
|
| 591 |
+
"transformer.layers.5.post_attention_layernorm.weight": "model-00015-of-00024.safetensors",
|
| 592 |
+
"transformer.layers.5.self_attn.k_proj.biases": "model-00015-of-00024.safetensors",
|
| 593 |
+
"transformer.layers.5.self_attn.k_proj.scales": "model-00015-of-00024.safetensors",
|
| 594 |
+
"transformer.layers.5.self_attn.k_proj.weight": "model-00015-of-00024.safetensors",
|
| 595 |
+
"transformer.layers.5.self_attn.o_proj.biases": "model-00015-of-00024.safetensors",
|
| 596 |
+
"transformer.layers.5.self_attn.o_proj.scales": "model-00015-of-00024.safetensors",
|
| 597 |
+
"transformer.layers.5.self_attn.o_proj.weight": "model-00015-of-00024.safetensors",
|
| 598 |
+
"transformer.layers.5.self_attn.q_proj.biases": "model-00015-of-00024.safetensors",
|
| 599 |
+
"transformer.layers.5.self_attn.q_proj.scales": "model-00015-of-00024.safetensors",
|
| 600 |
+
"transformer.layers.5.self_attn.q_proj.weight": "model-00016-of-00024.safetensors",
|
| 601 |
+
"transformer.layers.5.self_attn.v_proj.biases": "model-00016-of-00024.safetensors",
|
| 602 |
+
"transformer.layers.5.self_attn.v_proj.scales": "model-00016-of-00024.safetensors",
|
| 603 |
+
"transformer.layers.5.self_attn.v_proj.weight": "model-00016-of-00024.safetensors",
|
| 604 |
+
"transformer.layers.6.input_layernorm.weight": "model-00016-of-00024.safetensors",
|
| 605 |
+
"transformer.layers.6.mlp.down_proj.biases": "model-00016-of-00024.safetensors",
|
| 606 |
+
"transformer.layers.6.mlp.down_proj.scales": "model-00016-of-00024.safetensors",
|
| 607 |
+
"transformer.layers.6.mlp.down_proj.weight": "model-00016-of-00024.safetensors",
|
| 608 |
+
"transformer.layers.6.mlp.gate_proj.biases": "model-00016-of-00024.safetensors",
|
| 609 |
+
"transformer.layers.6.mlp.gate_proj.scales": "model-00016-of-00024.safetensors",
|
| 610 |
+
"transformer.layers.6.mlp.gate_proj.weight": "model-00016-of-00024.safetensors",
|
| 611 |
+
"transformer.layers.6.mlp.up_proj.biases": "model-00016-of-00024.safetensors",
|
| 612 |
+
"transformer.layers.6.mlp.up_proj.scales": "model-00016-of-00024.safetensors",
|
| 613 |
+
"transformer.layers.6.mlp.up_proj.weight": "model-00016-of-00024.safetensors",
|
| 614 |
+
"transformer.layers.6.post_attention_layernorm.weight": "model-00016-of-00024.safetensors",
|
| 615 |
+
"transformer.layers.6.self_attn.k_proj.biases": "model-00016-of-00024.safetensors",
|
| 616 |
+
"transformer.layers.6.self_attn.k_proj.scales": "model-00016-of-00024.safetensors",
|
| 617 |
+
"transformer.layers.6.self_attn.k_proj.weight": "model-00016-of-00024.safetensors",
|
| 618 |
+
"transformer.layers.6.self_attn.o_proj.biases": "model-00016-of-00024.safetensors",
|
| 619 |
+
"transformer.layers.6.self_attn.o_proj.scales": "model-00016-of-00024.safetensors",
|
| 620 |
+
"transformer.layers.6.self_attn.o_proj.weight": "model-00016-of-00024.safetensors",
|
| 621 |
+
"transformer.layers.6.self_attn.q_proj.biases": "model-00016-of-00024.safetensors",
|
| 622 |
+
"transformer.layers.6.self_attn.q_proj.scales": "model-00016-of-00024.safetensors",
|
| 623 |
+
"transformer.layers.6.self_attn.q_proj.weight": "model-00016-of-00024.safetensors",
|
| 624 |
+
"transformer.layers.6.self_attn.v_proj.biases": "model-00016-of-00024.safetensors",
|
| 625 |
+
"transformer.layers.6.self_attn.v_proj.scales": "model-00016-of-00024.safetensors",
|
| 626 |
+
"transformer.layers.6.self_attn.v_proj.weight": "model-00016-of-00024.safetensors",
|
| 627 |
+
"transformer.layers.7.input_layernorm.weight": "model-00016-of-00024.safetensors",
|
| 628 |
+
"transformer.layers.7.mlp.down_proj.biases": "model-00016-of-00024.safetensors",
|
| 629 |
+
"transformer.layers.7.mlp.down_proj.scales": "model-00016-of-00024.safetensors",
|
| 630 |
+
"transformer.layers.7.mlp.down_proj.weight": "model-00016-of-00024.safetensors",
|
| 631 |
+
"transformer.layers.7.mlp.gate_proj.biases": "model-00016-of-00024.safetensors",
|
| 632 |
+
"transformer.layers.7.mlp.gate_proj.scales": "model-00016-of-00024.safetensors",
|
| 633 |
+
"transformer.layers.7.mlp.gate_proj.weight": "model-00016-of-00024.safetensors",
|
| 634 |
+
"transformer.layers.7.mlp.up_proj.biases": "model-00016-of-00024.safetensors",
|
| 635 |
+
"transformer.layers.7.mlp.up_proj.scales": "model-00016-of-00024.safetensors",
|
| 636 |
+
"transformer.layers.7.mlp.up_proj.weight": "model-00016-of-00024.safetensors",
|
| 637 |
+
"transformer.layers.7.post_attention_layernorm.weight": "model-00016-of-00024.safetensors",
|
| 638 |
+
"transformer.layers.7.self_attn.k_proj.biases": "model-00016-of-00024.safetensors",
|
| 639 |
+
"transformer.layers.7.self_attn.k_proj.scales": "model-00016-of-00024.safetensors",
|
| 640 |
+
"transformer.layers.7.self_attn.k_proj.weight": "model-00016-of-00024.safetensors",
|
| 641 |
+
"transformer.layers.7.self_attn.o_proj.biases": "model-00016-of-00024.safetensors",
|
| 642 |
+
"transformer.layers.7.self_attn.o_proj.scales": "model-00016-of-00024.safetensors",
|
| 643 |
+
"transformer.layers.7.self_attn.o_proj.weight": "model-00016-of-00024.safetensors",
|
| 644 |
+
"transformer.layers.7.self_attn.q_proj.biases": "model-00016-of-00024.safetensors",
|
| 645 |
+
"transformer.layers.7.self_attn.q_proj.scales": "model-00016-of-00024.safetensors",
|
| 646 |
+
"transformer.layers.7.self_attn.q_proj.weight": "model-00016-of-00024.safetensors",
|
| 647 |
+
"transformer.layers.7.self_attn.v_proj.biases": "model-00016-of-00024.safetensors",
|
| 648 |
+
"transformer.layers.7.self_attn.v_proj.scales": "model-00016-of-00024.safetensors",
|
| 649 |
+
"transformer.layers.7.self_attn.v_proj.weight": "model-00016-of-00024.safetensors",
|
| 650 |
+
"transformer.layers.8.input_layernorm.weight": "model-00016-of-00024.safetensors",
|
| 651 |
+
"transformer.layers.8.mlp.down_proj.biases": "model-00016-of-00024.safetensors",
|
| 652 |
+
"transformer.layers.8.mlp.down_proj.scales": "model-00016-of-00024.safetensors",
|
| 653 |
+
"transformer.layers.8.mlp.down_proj.weight": "model-00017-of-00024.safetensors",
|
| 654 |
+
"transformer.layers.8.mlp.gate_proj.biases": "model-00017-of-00024.safetensors",
|
| 655 |
+
"transformer.layers.8.mlp.gate_proj.scales": "model-00017-of-00024.safetensors",
|
| 656 |
+
"transformer.layers.8.mlp.gate_proj.weight": "model-00017-of-00024.safetensors",
|
| 657 |
+
"transformer.layers.8.mlp.up_proj.biases": "model-00017-of-00024.safetensors",
|
| 658 |
+
"transformer.layers.8.mlp.up_proj.scales": "model-00017-of-00024.safetensors",
|
| 659 |
+
"transformer.layers.8.mlp.up_proj.weight": "model-00017-of-00024.safetensors",
|
| 660 |
+
"transformer.layers.8.post_attention_layernorm.weight": "model-00017-of-00024.safetensors",
|
| 661 |
+
"transformer.layers.8.self_attn.k_proj.biases": "model-00017-of-00024.safetensors",
|
| 662 |
+
"transformer.layers.8.self_attn.k_proj.scales": "model-00017-of-00024.safetensors",
|
| 663 |
+
"transformer.layers.8.self_attn.k_proj.weight": "model-00017-of-00024.safetensors",
|
| 664 |
+
"transformer.layers.8.self_attn.o_proj.biases": "model-00017-of-00024.safetensors",
|
| 665 |
+
"transformer.layers.8.self_attn.o_proj.scales": "model-00017-of-00024.safetensors",
|
| 666 |
+
"transformer.layers.8.self_attn.o_proj.weight": "model-00017-of-00024.safetensors",
|
| 667 |
+
"transformer.layers.8.self_attn.q_proj.biases": "model-00017-of-00024.safetensors",
|
| 668 |
+
"transformer.layers.8.self_attn.q_proj.scales": "model-00017-of-00024.safetensors",
|
| 669 |
+
"transformer.layers.8.self_attn.q_proj.weight": "model-00017-of-00024.safetensors",
|
| 670 |
+
"transformer.layers.8.self_attn.v_proj.biases": "model-00017-of-00024.safetensors",
|
| 671 |
+
"transformer.layers.8.self_attn.v_proj.scales": "model-00017-of-00024.safetensors",
|
| 672 |
+
"transformer.layers.8.self_attn.v_proj.weight": "model-00017-of-00024.safetensors",
|
| 673 |
+
"transformer.layers.9.input_layernorm.weight": "model-00017-of-00024.safetensors",
|
| 674 |
+
"transformer.layers.9.mlp.down_proj.biases": "model-00017-of-00024.safetensors",
|
| 675 |
+
"transformer.layers.9.mlp.down_proj.scales": "model-00017-of-00024.safetensors",
|
| 676 |
+
"transformer.layers.9.mlp.down_proj.weight": "model-00017-of-00024.safetensors",
|
| 677 |
+
"transformer.layers.9.mlp.gate_proj.biases": "model-00017-of-00024.safetensors",
|
| 678 |
+
"transformer.layers.9.mlp.gate_proj.scales": "model-00017-of-00024.safetensors",
|
| 679 |
+
"transformer.layers.9.mlp.gate_proj.weight": "model-00017-of-00024.safetensors",
|
| 680 |
+
"transformer.layers.9.mlp.up_proj.biases": "model-00017-of-00024.safetensors",
|
| 681 |
+
"transformer.layers.9.mlp.up_proj.scales": "model-00017-of-00024.safetensors",
|
| 682 |
+
"transformer.layers.9.mlp.up_proj.weight": "model-00017-of-00024.safetensors",
|
| 683 |
+
"transformer.layers.9.post_attention_layernorm.weight": "model-00017-of-00024.safetensors",
|
| 684 |
+
"transformer.layers.9.self_attn.k_proj.biases": "model-00017-of-00024.safetensors",
|
| 685 |
+
"transformer.layers.9.self_attn.k_proj.scales": "model-00017-of-00024.safetensors",
|
| 686 |
+
"transformer.layers.9.self_attn.k_proj.weight": "model-00017-of-00024.safetensors",
|
| 687 |
+
"transformer.layers.9.self_attn.o_proj.biases": "model-00017-of-00024.safetensors",
|
| 688 |
+
"transformer.layers.9.self_attn.o_proj.scales": "model-00017-of-00024.safetensors",
|
| 689 |
+
"transformer.layers.9.self_attn.o_proj.weight": "model-00017-of-00024.safetensors",
|
| 690 |
+
"transformer.layers.9.self_attn.q_proj.biases": "model-00017-of-00024.safetensors",
|
| 691 |
+
"transformer.layers.9.self_attn.q_proj.scales": "model-00017-of-00024.safetensors",
|
| 692 |
+
"transformer.layers.9.self_attn.q_proj.weight": "model-00017-of-00024.safetensors",
|
| 693 |
+
"transformer.layers.9.self_attn.v_proj.biases": "model-00017-of-00024.safetensors",
|
| 694 |
+
"transformer.layers.9.self_attn.v_proj.scales": "model-00017-of-00024.safetensors",
|
| 695 |
+
"transformer.layers.9.self_attn.v_proj.weight": "model-00017-of-00024.safetensors",
|
| 696 |
+
"transformer.norm.weight": "model-00017-of-00024.safetensors",
|
| 697 |
+
"transformer2.embed_tokens.biases": "model-00017-of-00024.safetensors",
|
| 698 |
+
"transformer2.embed_tokens.scales": "model-00017-of-00024.safetensors",
|
| 699 |
+
"transformer2.embed_tokens.weight": "model-00018-of-00024.safetensors",
|
| 700 |
+
"transformer2.layers.0.input_layernorm.weight": "model-00018-of-00024.safetensors",
|
| 701 |
+
"transformer2.layers.0.mlp.down_proj.biases": "model-00018-of-00024.safetensors",
|
| 702 |
+
"transformer2.layers.0.mlp.down_proj.scales": "model-00018-of-00024.safetensors",
|
| 703 |
+
"transformer2.layers.0.mlp.down_proj.weight": "model-00018-of-00024.safetensors",
|
| 704 |
+
"transformer2.layers.0.mlp.gate_proj.biases": "model-00018-of-00024.safetensors",
|
| 705 |
+
"transformer2.layers.0.mlp.gate_proj.scales": "model-00018-of-00024.safetensors",
|
| 706 |
+
"transformer2.layers.0.mlp.gate_proj.weight": "model-00018-of-00024.safetensors",
|
| 707 |
+
"transformer2.layers.0.mlp.up_proj.biases": "model-00018-of-00024.safetensors",
|
| 708 |
+
"transformer2.layers.0.mlp.up_proj.scales": "model-00018-of-00024.safetensors",
|
| 709 |
+
"transformer2.layers.0.mlp.up_proj.weight": "model-00018-of-00024.safetensors",
|
| 710 |
+
"transformer2.layers.0.post_attention_layernorm.weight": "model-00018-of-00024.safetensors",
|
| 711 |
+
"transformer2.layers.0.self_attn.k_proj.biases": "model-00018-of-00024.safetensors",
|
| 712 |
+
"transformer2.layers.0.self_attn.k_proj.scales": "model-00018-of-00024.safetensors",
|
| 713 |
+
"transformer2.layers.0.self_attn.k_proj.weight": "model-00018-of-00024.safetensors",
|
| 714 |
+
"transformer2.layers.0.self_attn.o_proj.biases": "model-00018-of-00024.safetensors",
|
| 715 |
+
"transformer2.layers.0.self_attn.o_proj.scales": "model-00018-of-00024.safetensors",
|
| 716 |
+
"transformer2.layers.0.self_attn.o_proj.weight": "model-00018-of-00024.safetensors",
|
| 717 |
+
"transformer2.layers.0.self_attn.q_proj.biases": "model-00018-of-00024.safetensors",
|
| 718 |
+
"transformer2.layers.0.self_attn.q_proj.scales": "model-00018-of-00024.safetensors",
|
| 719 |
+
"transformer2.layers.0.self_attn.q_proj.weight": "model-00018-of-00024.safetensors",
|
| 720 |
+
"transformer2.layers.0.self_attn.v_proj.biases": "model-00018-of-00024.safetensors",
|
| 721 |
+
"transformer2.layers.0.self_attn.v_proj.scales": "model-00018-of-00024.safetensors",
|
| 722 |
+
"transformer2.layers.0.self_attn.v_proj.weight": "model-00018-of-00024.safetensors",
|
| 723 |
+
"transformer2.layers.1.input_layernorm.weight": "model-00018-of-00024.safetensors",
|
| 724 |
+
"transformer2.layers.1.mlp.down_proj.biases": "model-00018-of-00024.safetensors",
|
| 725 |
+
"transformer2.layers.1.mlp.down_proj.scales": "model-00018-of-00024.safetensors",
|
| 726 |
+
"transformer2.layers.1.mlp.down_proj.weight": "model-00018-of-00024.safetensors",
|
| 727 |
+
"transformer2.layers.1.mlp.gate_proj.biases": "model-00018-of-00024.safetensors",
|
| 728 |
+
"transformer2.layers.1.mlp.gate_proj.scales": "model-00018-of-00024.safetensors",
|
| 729 |
+
"transformer2.layers.1.mlp.gate_proj.weight": "model-00018-of-00024.safetensors",
|
| 730 |
+
"transformer2.layers.1.mlp.up_proj.biases": "model-00018-of-00024.safetensors",
|
| 731 |
+
"transformer2.layers.1.mlp.up_proj.scales": "model-00018-of-00024.safetensors",
|
| 732 |
+
"transformer2.layers.1.mlp.up_proj.weight": "model-00018-of-00024.safetensors",
|
| 733 |
+
"transformer2.layers.1.post_attention_layernorm.weight": "model-00018-of-00024.safetensors",
|
| 734 |
+
"transformer2.layers.1.self_attn.k_proj.biases": "model-00018-of-00024.safetensors",
|
| 735 |
+
"transformer2.layers.1.self_attn.k_proj.scales": "model-00018-of-00024.safetensors",
|
| 736 |
+
"transformer2.layers.1.self_attn.k_proj.weight": "model-00018-of-00024.safetensors",
|
| 737 |
+
"transformer2.layers.1.self_attn.o_proj.biases": "model-00018-of-00024.safetensors",
|
| 738 |
+
"transformer2.layers.1.self_attn.o_proj.scales": "model-00018-of-00024.safetensors",
|
| 739 |
+
"transformer2.layers.1.self_attn.o_proj.weight": "model-00018-of-00024.safetensors",
|
| 740 |
+
"transformer2.layers.1.self_attn.q_proj.biases": "model-00018-of-00024.safetensors",
|
| 741 |
+
"transformer2.layers.1.self_attn.q_proj.scales": "model-00019-of-00024.safetensors",
|
| 742 |
+
"transformer2.layers.1.self_attn.q_proj.weight": "model-00019-of-00024.safetensors",
|
| 743 |
+
"transformer2.layers.1.self_attn.v_proj.biases": "model-00019-of-00024.safetensors",
|
| 744 |
+
"transformer2.layers.1.self_attn.v_proj.scales": "model-00019-of-00024.safetensors",
|
| 745 |
+
"transformer2.layers.1.self_attn.v_proj.weight": "model-00019-of-00024.safetensors",
|
| 746 |
+
"transformer2.layers.10.input_layernorm.weight": "model-00019-of-00024.safetensors",
|
| 747 |
+
"transformer2.layers.10.mlp.down_proj.biases": "model-00019-of-00024.safetensors",
|
| 748 |
+
"transformer2.layers.10.mlp.down_proj.scales": "model-00019-of-00024.safetensors",
|
| 749 |
+
"transformer2.layers.10.mlp.down_proj.weight": "model-00019-of-00024.safetensors",
|
| 750 |
+
"transformer2.layers.10.mlp.gate_proj.biases": "model-00019-of-00024.safetensors",
|
| 751 |
+
"transformer2.layers.10.mlp.gate_proj.scales": "model-00019-of-00024.safetensors",
|
| 752 |
+
"transformer2.layers.10.mlp.gate_proj.weight": "model-00019-of-00024.safetensors",
|
| 753 |
+
"transformer2.layers.10.mlp.up_proj.biases": "model-00019-of-00024.safetensors",
|
| 754 |
+
"transformer2.layers.10.mlp.up_proj.scales": "model-00019-of-00024.safetensors",
|
| 755 |
+
"transformer2.layers.10.mlp.up_proj.weight": "model-00019-of-00024.safetensors",
|
| 756 |
+
"transformer2.layers.10.post_attention_layernorm.weight": "model-00019-of-00024.safetensors",
|
| 757 |
+
"transformer2.layers.10.self_attn.k_proj.biases": "model-00019-of-00024.safetensors",
|
| 758 |
+
"transformer2.layers.10.self_attn.k_proj.scales": "model-00019-of-00024.safetensors",
|
| 759 |
+
"transformer2.layers.10.self_attn.k_proj.weight": "model-00019-of-00024.safetensors",
|
| 760 |
+
"transformer2.layers.10.self_attn.o_proj.biases": "model-00019-of-00024.safetensors",
|
| 761 |
+
"transformer2.layers.10.self_attn.o_proj.scales": "model-00019-of-00024.safetensors",
|
| 762 |
+
"transformer2.layers.10.self_attn.o_proj.weight": "model-00019-of-00024.safetensors",
|
| 763 |
+
"transformer2.layers.10.self_attn.q_proj.biases": "model-00019-of-00024.safetensors",
|
| 764 |
+
"transformer2.layers.10.self_attn.q_proj.scales": "model-00019-of-00024.safetensors",
|
| 765 |
+
"transformer2.layers.10.self_attn.q_proj.weight": "model-00019-of-00024.safetensors",
|
| 766 |
+
"transformer2.layers.10.self_attn.v_proj.biases": "model-00019-of-00024.safetensors",
|
| 767 |
+
"transformer2.layers.10.self_attn.v_proj.scales": "model-00019-of-00024.safetensors",
|
| 768 |
+
"transformer2.layers.10.self_attn.v_proj.weight": "model-00019-of-00024.safetensors",
|
| 769 |
+
"transformer2.layers.11.input_layernorm.weight": "model-00019-of-00024.safetensors",
|
| 770 |
+
"transformer2.layers.11.mlp.down_proj.biases": "model-00019-of-00024.safetensors",
|
| 771 |
+
"transformer2.layers.11.mlp.down_proj.scales": "model-00019-of-00024.safetensors",
|
| 772 |
+
"transformer2.layers.11.mlp.down_proj.weight": "model-00019-of-00024.safetensors",
|
| 773 |
+
"transformer2.layers.11.mlp.gate_proj.biases": "model-00019-of-00024.safetensors",
|
| 774 |
+
"transformer2.layers.11.mlp.gate_proj.scales": "model-00019-of-00024.safetensors",
|
| 775 |
+
"transformer2.layers.11.mlp.gate_proj.weight": "model-00019-of-00024.safetensors",
|
| 776 |
+
"transformer2.layers.11.mlp.up_proj.biases": "model-00019-of-00024.safetensors",
|
| 777 |
+
"transformer2.layers.11.mlp.up_proj.scales": "model-00019-of-00024.safetensors",
|
| 778 |
+
"transformer2.layers.11.mlp.up_proj.weight": "model-00019-of-00024.safetensors",
|
| 779 |
+
"transformer2.layers.11.post_attention_layernorm.weight": "model-00019-of-00024.safetensors",
|
| 780 |
+
"transformer2.layers.11.self_attn.k_proj.biases": "model-00019-of-00024.safetensors",
|
| 781 |
+
"transformer2.layers.11.self_attn.k_proj.scales": "model-00019-of-00024.safetensors",
|
| 782 |
+
"transformer2.layers.11.self_attn.k_proj.weight": "model-00019-of-00024.safetensors",
|
| 783 |
+
"transformer2.layers.11.self_attn.o_proj.biases": "model-00019-of-00024.safetensors",
|
| 784 |
+
"transformer2.layers.11.self_attn.o_proj.scales": "model-00019-of-00024.safetensors",
|
| 785 |
+
"transformer2.layers.11.self_attn.o_proj.weight": "model-00019-of-00024.safetensors",
|
| 786 |
+
"transformer2.layers.11.self_attn.q_proj.biases": "model-00019-of-00024.safetensors",
|
| 787 |
+
"transformer2.layers.11.self_attn.q_proj.scales": "model-00019-of-00024.safetensors",
|
| 788 |
+
"transformer2.layers.11.self_attn.q_proj.weight": "model-00019-of-00024.safetensors",
|
| 789 |
+
"transformer2.layers.11.self_attn.v_proj.biases": "model-00019-of-00024.safetensors",
|
| 790 |
+
"transformer2.layers.11.self_attn.v_proj.scales": "model-00019-of-00024.safetensors",
|
| 791 |
+
"transformer2.layers.11.self_attn.v_proj.weight": "model-00019-of-00024.safetensors",
|
| 792 |
+
"transformer2.layers.2.input_layernorm.weight": "model-00019-of-00024.safetensors",
|
| 793 |
+
"transformer2.layers.2.mlp.down_proj.biases": "model-00019-of-00024.safetensors",
|
| 794 |
+
"transformer2.layers.2.mlp.down_proj.scales": "model-00019-of-00024.safetensors",
|
| 795 |
+
"transformer2.layers.2.mlp.down_proj.weight": "model-00020-of-00024.safetensors",
|
| 796 |
+
"transformer2.layers.2.mlp.gate_proj.biases": "model-00020-of-00024.safetensors",
|
| 797 |
+
"transformer2.layers.2.mlp.gate_proj.scales": "model-00020-of-00024.safetensors",
|
| 798 |
+
"transformer2.layers.2.mlp.gate_proj.weight": "model-00020-of-00024.safetensors",
|
| 799 |
+
"transformer2.layers.2.mlp.up_proj.biases": "model-00020-of-00024.safetensors",
|
| 800 |
+
"transformer2.layers.2.mlp.up_proj.scales": "model-00020-of-00024.safetensors",
|
| 801 |
+
"transformer2.layers.2.mlp.up_proj.weight": "model-00020-of-00024.safetensors",
|
| 802 |
+
"transformer2.layers.2.post_attention_layernorm.weight": "model-00020-of-00024.safetensors",
|
| 803 |
+
"transformer2.layers.2.self_attn.k_proj.biases": "model-00020-of-00024.safetensors",
|
| 804 |
+
"transformer2.layers.2.self_attn.k_proj.scales": "model-00020-of-00024.safetensors",
|
| 805 |
+
"transformer2.layers.2.self_attn.k_proj.weight": "model-00020-of-00024.safetensors",
|
| 806 |
+
"transformer2.layers.2.self_attn.o_proj.biases": "model-00020-of-00024.safetensors",
|
| 807 |
+
"transformer2.layers.2.self_attn.o_proj.scales": "model-00020-of-00024.safetensors",
|
| 808 |
+
"transformer2.layers.2.self_attn.o_proj.weight": "model-00020-of-00024.safetensors",
|
| 809 |
+
"transformer2.layers.2.self_attn.q_proj.biases": "model-00020-of-00024.safetensors",
|
| 810 |
+
"transformer2.layers.2.self_attn.q_proj.scales": "model-00020-of-00024.safetensors",
|
| 811 |
+
"transformer2.layers.2.self_attn.q_proj.weight": "model-00020-of-00024.safetensors",
|
| 812 |
+
"transformer2.layers.2.self_attn.v_proj.biases": "model-00020-of-00024.safetensors",
|
| 813 |
+
"transformer2.layers.2.self_attn.v_proj.scales": "model-00020-of-00024.safetensors",
|
| 814 |
+
"transformer2.layers.2.self_attn.v_proj.weight": "model-00020-of-00024.safetensors",
|
| 815 |
+
"transformer2.layers.3.input_layernorm.weight": "model-00020-of-00024.safetensors",
|
| 816 |
+
"transformer2.layers.3.mlp.down_proj.biases": "model-00020-of-00024.safetensors",
|
| 817 |
+
"transformer2.layers.3.mlp.down_proj.scales": "model-00020-of-00024.safetensors",
|
| 818 |
+
"transformer2.layers.3.mlp.down_proj.weight": "model-00020-of-00024.safetensors",
|
| 819 |
+
"transformer2.layers.3.mlp.gate_proj.biases": "model-00020-of-00024.safetensors",
|
| 820 |
+
"transformer2.layers.3.mlp.gate_proj.scales": "model-00020-of-00024.safetensors",
|
| 821 |
+
"transformer2.layers.3.mlp.gate_proj.weight": "model-00020-of-00024.safetensors",
|
| 822 |
+
"transformer2.layers.3.mlp.up_proj.biases": "model-00020-of-00024.safetensors",
|
| 823 |
+
"transformer2.layers.3.mlp.up_proj.scales": "model-00020-of-00024.safetensors",
|
| 824 |
+
"transformer2.layers.3.mlp.up_proj.weight": "model-00020-of-00024.safetensors",
|
| 825 |
+
"transformer2.layers.3.post_attention_layernorm.weight": "model-00020-of-00024.safetensors",
|
| 826 |
+
"transformer2.layers.3.self_attn.k_proj.biases": "model-00020-of-00024.safetensors",
|
| 827 |
+
"transformer2.layers.3.self_attn.k_proj.scales": "model-00020-of-00024.safetensors",
|
| 828 |
+
"transformer2.layers.3.self_attn.k_proj.weight": "model-00020-of-00024.safetensors",
|
| 829 |
+
"transformer2.layers.3.self_attn.o_proj.biases": "model-00020-of-00024.safetensors",
|
| 830 |
+
"transformer2.layers.3.self_attn.o_proj.scales": "model-00020-of-00024.safetensors",
|
| 831 |
+
"transformer2.layers.3.self_attn.o_proj.weight": "model-00020-of-00024.safetensors",
|
| 832 |
+
"transformer2.layers.3.self_attn.q_proj.biases": "model-00020-of-00024.safetensors",
|
| 833 |
+
"transformer2.layers.3.self_attn.q_proj.scales": "model-00020-of-00024.safetensors",
|
| 834 |
+
"transformer2.layers.3.self_attn.q_proj.weight": "model-00020-of-00024.safetensors",
|
| 835 |
+
"transformer2.layers.3.self_attn.v_proj.biases": "model-00020-of-00024.safetensors",
|
| 836 |
+
"transformer2.layers.3.self_attn.v_proj.scales": "model-00020-of-00024.safetensors",
|
| 837 |
+
"transformer2.layers.3.self_attn.v_proj.weight": "model-00020-of-00024.safetensors",
|
| 838 |
+
"transformer2.layers.4.input_layernorm.weight": "model-00020-of-00024.safetensors",
|
| 839 |
+
"transformer2.layers.4.mlp.down_proj.biases": "model-00020-of-00024.safetensors",
|
| 840 |
+
"transformer2.layers.4.mlp.down_proj.scales": "model-00020-of-00024.safetensors",
|
| 841 |
+
"transformer2.layers.4.mlp.down_proj.weight": "model-00020-of-00024.safetensors",
|
| 842 |
+
"transformer2.layers.4.mlp.gate_proj.biases": "model-00020-of-00024.safetensors",
|
| 843 |
+
"transformer2.layers.4.mlp.gate_proj.scales": "model-00020-of-00024.safetensors",
|
| 844 |
+
"transformer2.layers.4.mlp.gate_proj.weight": "model-00021-of-00024.safetensors",
|
| 845 |
+
"transformer2.layers.4.mlp.up_proj.biases": "model-00021-of-00024.safetensors",
|
| 846 |
+
"transformer2.layers.4.mlp.up_proj.scales": "model-00021-of-00024.safetensors",
|
| 847 |
+
"transformer2.layers.4.mlp.up_proj.weight": "model-00021-of-00024.safetensors",
|
| 848 |
+
"transformer2.layers.4.post_attention_layernorm.weight": "model-00021-of-00024.safetensors",
|
| 849 |
+
"transformer2.layers.4.self_attn.k_proj.biases": "model-00021-of-00024.safetensors",
|
| 850 |
+
"transformer2.layers.4.self_attn.k_proj.scales": "model-00021-of-00024.safetensors",
|
| 851 |
+
"transformer2.layers.4.self_attn.k_proj.weight": "model-00021-of-00024.safetensors",
|
| 852 |
+
"transformer2.layers.4.self_attn.o_proj.biases": "model-00021-of-00024.safetensors",
|
| 853 |
+
"transformer2.layers.4.self_attn.o_proj.scales": "model-00021-of-00024.safetensors",
|
| 854 |
+
"transformer2.layers.4.self_attn.o_proj.weight": "model-00021-of-00024.safetensors",
|
| 855 |
+
"transformer2.layers.4.self_attn.q_proj.biases": "model-00021-of-00024.safetensors",
|
| 856 |
+
"transformer2.layers.4.self_attn.q_proj.scales": "model-00021-of-00024.safetensors",
|
| 857 |
+
"transformer2.layers.4.self_attn.q_proj.weight": "model-00021-of-00024.safetensors",
|
| 858 |
+
"transformer2.layers.4.self_attn.v_proj.biases": "model-00021-of-00024.safetensors",
|
| 859 |
+
"transformer2.layers.4.self_attn.v_proj.scales": "model-00021-of-00024.safetensors",
|
| 860 |
+
"transformer2.layers.4.self_attn.v_proj.weight": "model-00021-of-00024.safetensors",
|
| 861 |
+
"transformer2.layers.5.input_layernorm.weight": "model-00021-of-00024.safetensors",
|
| 862 |
+
"transformer2.layers.5.mlp.down_proj.biases": "model-00021-of-00024.safetensors",
|
| 863 |
+
"transformer2.layers.5.mlp.down_proj.scales": "model-00021-of-00024.safetensors",
|
| 864 |
+
"transformer2.layers.5.mlp.down_proj.weight": "model-00021-of-00024.safetensors",
|
| 865 |
+
"transformer2.layers.5.mlp.gate_proj.biases": "model-00021-of-00024.safetensors",
|
| 866 |
+
"transformer2.layers.5.mlp.gate_proj.scales": "model-00021-of-00024.safetensors",
|
| 867 |
+
"transformer2.layers.5.mlp.gate_proj.weight": "model-00021-of-00024.safetensors",
|
| 868 |
+
"transformer2.layers.5.mlp.up_proj.biases": "model-00021-of-00024.safetensors",
|
| 869 |
+
"transformer2.layers.5.mlp.up_proj.scales": "model-00021-of-00024.safetensors",
|
| 870 |
+
"transformer2.layers.5.mlp.up_proj.weight": "model-00021-of-00024.safetensors",
|
| 871 |
+
"transformer2.layers.5.post_attention_layernorm.weight": "model-00021-of-00024.safetensors",
|
| 872 |
+
"transformer2.layers.5.self_attn.k_proj.biases": "model-00021-of-00024.safetensors",
|
| 873 |
+
"transformer2.layers.5.self_attn.k_proj.scales": "model-00021-of-00024.safetensors",
|
| 874 |
+
"transformer2.layers.5.self_attn.k_proj.weight": "model-00021-of-00024.safetensors",
|
| 875 |
+
"transformer2.layers.5.self_attn.o_proj.biases": "model-00021-of-00024.safetensors",
|
| 876 |
+
"transformer2.layers.5.self_attn.o_proj.scales": "model-00021-of-00024.safetensors",
|
| 877 |
+
"transformer2.layers.5.self_attn.o_proj.weight": "model-00021-of-00024.safetensors",
|
| 878 |
+
"transformer2.layers.5.self_attn.q_proj.biases": "model-00021-of-00024.safetensors",
|
| 879 |
+
"transformer2.layers.5.self_attn.q_proj.scales": "model-00021-of-00024.safetensors",
|
| 880 |
+
"transformer2.layers.5.self_attn.q_proj.weight": "model-00021-of-00024.safetensors",
|
| 881 |
+
"transformer2.layers.5.self_attn.v_proj.biases": "model-00021-of-00024.safetensors",
|
| 882 |
+
"transformer2.layers.5.self_attn.v_proj.scales": "model-00021-of-00024.safetensors",
|
| 883 |
+
"transformer2.layers.5.self_attn.v_proj.weight": "model-00021-of-00024.safetensors",
|
| 884 |
+
"transformer2.layers.6.input_layernorm.weight": "model-00021-of-00024.safetensors",
|
| 885 |
+
"transformer2.layers.6.mlp.down_proj.biases": "model-00021-of-00024.safetensors",
|
| 886 |
+
"transformer2.layers.6.mlp.down_proj.scales": "model-00021-of-00024.safetensors",
|
| 887 |
+
"transformer2.layers.6.mlp.down_proj.weight": "model-00021-of-00024.safetensors",
|
| 888 |
+
"transformer2.layers.6.mlp.gate_proj.biases": "model-00021-of-00024.safetensors",
|
| 889 |
+
"transformer2.layers.6.mlp.gate_proj.scales": "model-00021-of-00024.safetensors",
|
| 890 |
+
"transformer2.layers.6.mlp.gate_proj.weight": "model-00021-of-00024.safetensors",
|
| 891 |
+
"transformer2.layers.6.mlp.up_proj.biases": "model-00021-of-00024.safetensors",
|
| 892 |
+
"transformer2.layers.6.mlp.up_proj.scales": "model-00021-of-00024.safetensors",
|
| 893 |
+
"transformer2.layers.6.mlp.up_proj.weight": "model-00022-of-00024.safetensors",
|
| 894 |
+
"transformer2.layers.6.post_attention_layernorm.weight": "model-00022-of-00024.safetensors",
|
| 895 |
+
"transformer2.layers.6.self_attn.k_proj.biases": "model-00022-of-00024.safetensors",
|
| 896 |
+
"transformer2.layers.6.self_attn.k_proj.scales": "model-00022-of-00024.safetensors",
|
| 897 |
+
"transformer2.layers.6.self_attn.k_proj.weight": "model-00022-of-00024.safetensors",
|
| 898 |
+
"transformer2.layers.6.self_attn.o_proj.biases": "model-00022-of-00024.safetensors",
|
| 899 |
+
"transformer2.layers.6.self_attn.o_proj.scales": "model-00022-of-00024.safetensors",
|
| 900 |
+
"transformer2.layers.6.self_attn.o_proj.weight": "model-00022-of-00024.safetensors",
|
| 901 |
+
"transformer2.layers.6.self_attn.q_proj.biases": "model-00022-of-00024.safetensors",
|
| 902 |
+
"transformer2.layers.6.self_attn.q_proj.scales": "model-00022-of-00024.safetensors",
|
| 903 |
+
"transformer2.layers.6.self_attn.q_proj.weight": "model-00022-of-00024.safetensors",
|
| 904 |
+
"transformer2.layers.6.self_attn.v_proj.biases": "model-00022-of-00024.safetensors",
|
| 905 |
+
"transformer2.layers.6.self_attn.v_proj.scales": "model-00022-of-00024.safetensors",
|
| 906 |
+
"transformer2.layers.6.self_attn.v_proj.weight": "model-00022-of-00024.safetensors",
|
| 907 |
+
"transformer2.layers.7.input_layernorm.weight": "model-00022-of-00024.safetensors",
|
| 908 |
+
"transformer2.layers.7.mlp.down_proj.biases": "model-00022-of-00024.safetensors",
|
| 909 |
+
"transformer2.layers.7.mlp.down_proj.scales": "model-00022-of-00024.safetensors",
|
| 910 |
+
"transformer2.layers.7.mlp.down_proj.weight": "model-00022-of-00024.safetensors",
|
| 911 |
+
"transformer2.layers.7.mlp.gate_proj.biases": "model-00022-of-00024.safetensors",
|
| 912 |
+
"transformer2.layers.7.mlp.gate_proj.scales": "model-00022-of-00024.safetensors",
|
| 913 |
+
"transformer2.layers.7.mlp.gate_proj.weight": "model-00022-of-00024.safetensors",
|
| 914 |
+
"transformer2.layers.7.mlp.up_proj.biases": "model-00022-of-00024.safetensors",
|
| 915 |
+
"transformer2.layers.7.mlp.up_proj.scales": "model-00022-of-00024.safetensors",
|
| 916 |
+
"transformer2.layers.7.mlp.up_proj.weight": "model-00022-of-00024.safetensors",
|
| 917 |
+
"transformer2.layers.7.post_attention_layernorm.weight": "model-00022-of-00024.safetensors",
|
| 918 |
+
"transformer2.layers.7.self_attn.k_proj.biases": "model-00022-of-00024.safetensors",
|
| 919 |
+
"transformer2.layers.7.self_attn.k_proj.scales": "model-00022-of-00024.safetensors",
|
| 920 |
+
"transformer2.layers.7.self_attn.k_proj.weight": "model-00022-of-00024.safetensors",
|
| 921 |
+
"transformer2.layers.7.self_attn.o_proj.biases": "model-00022-of-00024.safetensors",
|
| 922 |
+
"transformer2.layers.7.self_attn.o_proj.scales": "model-00022-of-00024.safetensors",
|
| 923 |
+
"transformer2.layers.7.self_attn.o_proj.weight": "model-00022-of-00024.safetensors",
|
| 924 |
+
"transformer2.layers.7.self_attn.q_proj.biases": "model-00022-of-00024.safetensors",
|
| 925 |
+
"transformer2.layers.7.self_attn.q_proj.scales": "model-00022-of-00024.safetensors",
|
| 926 |
+
"transformer2.layers.7.self_attn.q_proj.weight": "model-00022-of-00024.safetensors",
|
| 927 |
+
"transformer2.layers.7.self_attn.v_proj.biases": "model-00022-of-00024.safetensors",
|
| 928 |
+
"transformer2.layers.7.self_attn.v_proj.scales": "model-00022-of-00024.safetensors",
|
| 929 |
+
"transformer2.layers.7.self_attn.v_proj.weight": "model-00022-of-00024.safetensors",
|
| 930 |
+
"transformer2.layers.8.input_layernorm.weight": "model-00022-of-00024.safetensors",
|
| 931 |
+
"transformer2.layers.8.mlp.down_proj.biases": "model-00022-of-00024.safetensors",
|
| 932 |
+
"transformer2.layers.8.mlp.down_proj.scales": "model-00022-of-00024.safetensors",
|
| 933 |
+
"transformer2.layers.8.mlp.down_proj.weight": "model-00022-of-00024.safetensors",
|
| 934 |
+
"transformer2.layers.8.mlp.gate_proj.biases": "model-00022-of-00024.safetensors",
|
| 935 |
+
"transformer2.layers.8.mlp.gate_proj.scales": "model-00022-of-00024.safetensors",
|
| 936 |
+
"transformer2.layers.8.mlp.gate_proj.weight": "model-00022-of-00024.safetensors",
|
| 937 |
+
"transformer2.layers.8.mlp.up_proj.biases": "model-00022-of-00024.safetensors",
|
| 938 |
+
"transformer2.layers.8.mlp.up_proj.scales": "model-00022-of-00024.safetensors",
|
| 939 |
+
"transformer2.layers.8.mlp.up_proj.weight": "model-00022-of-00024.safetensors",
|
| 940 |
+
"transformer2.layers.8.post_attention_layernorm.weight": "model-00022-of-00024.safetensors",
|
| 941 |
+
"transformer2.layers.8.self_attn.k_proj.biases": "model-00022-of-00024.safetensors",
|
| 942 |
+
"transformer2.layers.8.self_attn.k_proj.scales": "model-00022-of-00024.safetensors",
|
| 943 |
+
"transformer2.layers.8.self_attn.k_proj.weight": "model-00022-of-00024.safetensors",
|
| 944 |
+
"transformer2.layers.8.self_attn.o_proj.biases": "model-00022-of-00024.safetensors",
|
| 945 |
+
"transformer2.layers.8.self_attn.o_proj.scales": "model-00022-of-00024.safetensors",
|
| 946 |
+
"transformer2.layers.8.self_attn.o_proj.weight": "model-00022-of-00024.safetensors",
|
| 947 |
+
"transformer2.layers.8.self_attn.q_proj.biases": "model-00022-of-00024.safetensors",
|
| 948 |
+
"transformer2.layers.8.self_attn.q_proj.scales": "model-00022-of-00024.safetensors",
|
| 949 |
+
"transformer2.layers.8.self_attn.q_proj.weight": "model-00023-of-00024.safetensors",
|
| 950 |
+
"transformer2.layers.8.self_attn.v_proj.biases": "model-00023-of-00024.safetensors",
|
| 951 |
+
"transformer2.layers.8.self_attn.v_proj.scales": "model-00023-of-00024.safetensors",
|
| 952 |
+
"transformer2.layers.8.self_attn.v_proj.weight": "model-00023-of-00024.safetensors",
|
| 953 |
+
"transformer2.layers.9.input_layernorm.weight": "model-00023-of-00024.safetensors",
|
| 954 |
+
"transformer2.layers.9.mlp.down_proj.biases": "model-00023-of-00024.safetensors",
|
| 955 |
+
"transformer2.layers.9.mlp.down_proj.scales": "model-00023-of-00024.safetensors",
|
| 956 |
+
"transformer2.layers.9.mlp.down_proj.weight": "model-00023-of-00024.safetensors",
|
| 957 |
+
"transformer2.layers.9.mlp.gate_proj.biases": "model-00023-of-00024.safetensors",
|
| 958 |
+
"transformer2.layers.9.mlp.gate_proj.scales": "model-00023-of-00024.safetensors",
|
| 959 |
+
"transformer2.layers.9.mlp.gate_proj.weight": "model-00023-of-00024.safetensors",
|
| 960 |
+
"transformer2.layers.9.mlp.up_proj.biases": "model-00023-of-00024.safetensors",
|
| 961 |
+
"transformer2.layers.9.mlp.up_proj.scales": "model-00023-of-00024.safetensors",
|
| 962 |
+
"transformer2.layers.9.mlp.up_proj.weight": "model-00023-of-00024.safetensors",
|
| 963 |
+
"transformer2.layers.9.post_attention_layernorm.weight": "model-00023-of-00024.safetensors",
|
| 964 |
+
"transformer2.layers.9.self_attn.k_proj.biases": "model-00023-of-00024.safetensors",
|
| 965 |
+
"transformer2.layers.9.self_attn.k_proj.scales": "model-00023-of-00024.safetensors",
|
| 966 |
+
"transformer2.layers.9.self_attn.k_proj.weight": "model-00023-of-00024.safetensors",
|
| 967 |
+
"transformer2.layers.9.self_attn.o_proj.biases": "model-00023-of-00024.safetensors",
|
| 968 |
+
"transformer2.layers.9.self_attn.o_proj.scales": "model-00023-of-00024.safetensors",
|
| 969 |
+
"transformer2.layers.9.self_attn.o_proj.weight": "model-00023-of-00024.safetensors",
|
| 970 |
+
"transformer2.layers.9.self_attn.q_proj.biases": "model-00023-of-00024.safetensors",
|
| 971 |
+
"transformer2.layers.9.self_attn.q_proj.scales": "model-00023-of-00024.safetensors",
|
| 972 |
+
"transformer2.layers.9.self_attn.q_proj.weight": "model-00023-of-00024.safetensors",
|
| 973 |
+
"transformer2.layers.9.self_attn.v_proj.biases": "model-00023-of-00024.safetensors",
|
| 974 |
+
"transformer2.layers.9.self_attn.v_proj.scales": "model-00023-of-00024.safetensors",
|
| 975 |
+
"transformer2.layers.9.self_attn.v_proj.weight": "model-00023-of-00024.safetensors",
|
| 976 |
+
"transformer2.norm.weight": "model-00023-of-00024.safetensors",
|
| 977 |
+
"transformer_lm_head.biases": "model-00023-of-00024.safetensors",
|
| 978 |
+
"transformer_lm_head.scales": "model-00023-of-00024.safetensors",
|
| 979 |
+
"transformer_lm_head.weight": "model-00023-of-00024.safetensors",
|
| 980 |
+
"type_info_output_proj.biases": "model-00023-of-00024.safetensors",
|
| 981 |
+
"type_info_output_proj.scales": "model-00023-of-00024.safetensors",
|
| 982 |
+
"type_info_output_proj.weight": "model-00024-of-00024.safetensors"
|
| 983 |
+
}
|
| 984 |
+
}
|
qwen2_tokenizer/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen2ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_dropout": 0.0,
|
| 6 |
+
"bos_token_id": 151643,
|
| 7 |
+
"eos_token_id": 151643,
|
| 8 |
+
"hidden_act": "silu",
|
| 9 |
+
"hidden_size": 3584,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 18944,
|
| 12 |
+
"max_position_embeddings": 131072,
|
| 13 |
+
"max_window_layers": 28,
|
| 14 |
+
"model_type": "qwen2",
|
| 15 |
+
"num_attention_heads": 28,
|
| 16 |
+
"num_hidden_layers": 28,
|
| 17 |
+
"num_key_value_heads": 4,
|
| 18 |
+
"rms_norm_eps": 1e-06,
|
| 19 |
+
"rope_theta": 1000000.0,
|
| 20 |
+
"sliding_window": 131072,
|
| 21 |
+
"tie_word_embeddings": false,
|
| 22 |
+
"torch_dtype": "bfloat16",
|
| 23 |
+
"transformers_version": "4.37.2",
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"use_sliding_window": false,
|
| 26 |
+
"vocab_size": 152064
|
| 27 |
+
}
|
qwen2_tokenizer/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": false,
|
| 4 |
+
"eos_token_id": 151643,
|
| 5 |
+
"max_new_tokens": 2048,
|
| 6 |
+
"transformers_version": "4.37.0"
|
| 7 |
+
}
|
qwen2_tokenizer/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qwen2_tokenizer/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
qwen2_tokenizer/tokenizer_config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"151643": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"151644": {
|
| 13 |
+
"content": "<|im_start|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"151645": {
|
| 21 |
+
"content": "<|im_end|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"additional_special_tokens": ["<|im_start|>", "<|im_end|>"],
|
| 30 |
+
"bos_token": null,
|
| 31 |
+
"chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 32 |
+
"clean_up_tokenization_spaces": false,
|
| 33 |
+
"eos_token": "<|endoftext|>",
|
| 34 |
+
"errors": "replace",
|
| 35 |
+
"model_max_length": 32768,
|
| 36 |
+
"pad_token": "<|endoftext|>",
|
| 37 |
+
"split_special_tokens": false,
|
| 38 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 39 |
+
"unk_token": null
|
| 40 |
+
}
|
qwen2_tokenizer/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
vocab.yaml
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
- '[verse]'
|
| 2 |
+
- '[chorus]'
|
| 3 |
+
- '[bridge]'
|
| 4 |
+
- '[intro-short]'
|
| 5 |
+
- '[intro-medium]'
|
| 6 |
+
- '[intro-long]'
|
| 7 |
+
- '[outro-short]'
|
| 8 |
+
- '[outro-medium]'
|
| 9 |
+
- '[outro-long]'
|
| 10 |
+
- '[inst-short]'
|
| 11 |
+
- '[inst-medium]'
|
| 12 |
+
- '[inst-long]'
|
| 13 |
+
- '[silence]'
|