ssbtech commited on
Commit
c6e2fc0
·
verified ·
1 Parent(s): 53485ac

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -33,13 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
- samples/af_heart_3.wav filter=lfs diff=lfs merge=lfs -text
37
- samples/af_heart_4.wav filter=lfs diff=lfs merge=lfs -text
38
- samples/af_heart_5.wav filter=lfs diff=lfs merge=lfs -text
39
- eval/ArtificialAnalysis-2025-02-26.jpeg filter=lfs diff=lfs merge=lfs -text
40
- eval/TTS_Arena-2025-02-26.jpeg filter=lfs diff=lfs merge=lfs -text
41
- eval/TTS_Spaces_Arena-2025-02-26.jpeg filter=lfs diff=lfs merge=lfs -text
42
- samples/HEARME.wav filter=lfs diff=lfs merge=lfs -text
43
- samples/af_heart_0.wav filter=lfs diff=lfs merge=lfs -text
44
- samples/af_heart_1.wav filter=lfs diff=lfs merge=lfs -text
45
- samples/af_heart_2.wav filter=lfs diff=lfs merge=lfs -text
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ assets/logo.png filter=lfs diff=lfs merge=lfs -text
37
+ assets/logo2.jpeg filter=lfs diff=lfs merge=lfs -text
38
+ assets/pipe.png filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
README.md CHANGED
@@ -2,116 +2,87 @@
2
  license: apache-2.0
3
  language:
4
  - en
5
- base_model:
6
- - yl4579/StyleTTS2-LJSpeech
7
- pipeline_tag: text-to-speech
 
 
 
8
  ---
9
- **Kokoro** is an open-weight TTS model with 82 million parameters. Despite its lightweight architecture, it delivers comparable quality to larger models while being significantly faster and more cost-efficient. With Apache-licensed weights, Kokoro can be deployed anywhere from production environments to personal projects.
10
-
11
- <audio controls><source src="https://huggingface.co/hexgrad/Kokoro-82M/resolve/main/samples/HEARME.wav" type="audio/wav"></audio>
12
-
13
- 🐈 **GitHub**: https://github.com/hexgrad/kokoro
14
-
15
- 🚀 **Demo**: https://hf.co/spaces/hexgrad/Kokoro-TTS
16
-
17
- > [!NOTE]
18
- > As of April 2025, the market rate of Kokoro served over API is **under $1 per million characters of text input**, or under $0.06 per hour of audio output. (On average, 1000 characters of input is about 1 minute of output.) Sources: [ArtificialAnalysis/Replicate at 65 cents per M chars](https://artificialanalysis.ai/text-to-speech/model-family/kokoro#price) and [DeepInfra at 80 cents per M chars](https://deepinfra.com/hexgrad/Kokoro-82M).
19
- >
20
- > This is an Apache-licensed model, and Kokoro has been deployed in numerous projects and commercial APIs. We welcome the deployment of the model in real use cases.
21
-
22
- > [!CAUTION]
23
- > Fake websites like kokorottsai_com (snapshot: https://archive.ph/nRRnk) and kokorotts_net (snapshot: https://archive.ph/60opa) are likely scams masquerading under the banner of a popular model.
24
- >
25
- > Any website containing "kokoro" in its root domain (e.g. kokorottsai_com, kokorotts_net) is **NOT owned by and NOT affiliated with this model page or its author**, and attempts to imply otherwise are red flags.
26
-
27
- - [Releases](#releases)
28
- - [Usage](#usage)
29
- - [EVAL.md](https://huggingface.co/hexgrad/Kokoro-82M/blob/main/EVAL.md) ↗️
30
- - [SAMPLES.md](https://huggingface.co/hexgrad/Kokoro-82M/blob/main/SAMPLES.md) ↗️
31
- - [VOICES.md](https://huggingface.co/hexgrad/Kokoro-82M/blob/main/VOICES.md) ↗️
32
- - [Model Facts](#model-facts)
33
- - [Training Details](#training-details)
34
- - [Creative Commons Attribution](#creative-commons-attribution)
35
- - [Acknowledgements](#acknowledgements)
36
-
37
- ### Releases
38
-
39
- | Model | Published | Training Data | Langs & Voices | SHA256 |
40
- | ----- | --------- | ------------- | -------------- | ------ |
41
- | **v1.0** | **2025 Jan 27** | **Few hundred hrs** | [**8 & 54**](https://huggingface.co/hexgrad/Kokoro-82M/blob/main/VOICES.md) | `496dba11` |
42
- | [v0.19](https://huggingface.co/hexgrad/kLegacy/tree/main/v0.19) | 2024 Dec 25 | <100 hrs | 1 & 10 | `3b0c392f` |
43
-
44
- | Training Costs | v0.19 | v1.0 | **Total** |
45
- | -------------- | ----- | ---- | ----- |
46
- | in A100 80GB GPU hours | 500 | 500 | **1000** |
47
- | average hourly rate | $0.80/h | $1.20/h | **$1/h** |
48
- | in USD | $400 | $600 | **$1000** |
49
-
50
- ### Usage
51
- You can run this basic cell on [Google Colab](https://colab.research.google.com/). [Listen to samples](https://huggingface.co/hexgrad/Kokoro-82M/blob/main/SAMPLES.md). For more languages and details, see [Advanced Usage](https://github.com/hexgrad/kokoro?tab=readme-ov-file#advanced-usage).
52
- ```py
53
- !pip install -q kokoro>=0.9.2 soundfile
54
- !apt-get -qq -y install espeak-ng > /dev/null 2>&1
55
- from kokoro import KPipeline
56
- from IPython.display import display, Audio
57
- import soundfile as sf
58
- import torch
59
- pipeline = KPipeline(lang_code='a')
60
- text = '''
61
- [Kokoro](/kˈOkəɹO/) is an open-weight TTS model with 82 million parameters. Despite its lightweight architecture, it delivers comparable quality to larger models while being significantly faster and more cost-efficient. With Apache-licensed weights, [Kokoro](/kˈOkəɹO/) can be deployed anywhere from production environments to personal projects.
62
- '''
63
- generator = pipeline(text, voice='af_heart')
64
- for i, (gs, ps, audio) in enumerate(generator):
65
- print(i, gs, ps)
66
- display(Audio(data=audio, rate=24000, autoplay=i==0))
67
- sf.write(f'{i}.wav', audio, 24000)
68
- ```
69
- Under the hood, `kokoro` uses [`misaki`](https://pypi.org/project/misaki/), a G2P library at https://github.com/hexgrad/misaki
70
 
71
- ### Model Facts
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
72
 
73
- **Architecture:**
74
- - StyleTTS 2: https://arxiv.org/abs/2306.07691
75
- - ISTFTNet: https://arxiv.org/abs/2203.02395
76
- - Decoder only: no diffusion, no encoder release
77
 
78
- **Architected by:** Li et al @ https://github.com/yl4579/StyleTTS2
79
 
80
- **Trained by**: `@rzvzn` on Discord
81
 
82
- **Languages:** Multiple
83
 
84
- **Model SHA256 Hash:** `496dba118d1a58f5f3db2efc88dbdc216e0483fc89fe6e47ee1f2c53f18ad1e4`
85
 
86
- ### Training Details
87
 
88
- **Data:** Kokoro was trained exclusively on **permissive/non-copyrighted audio data** and IPA phoneme labels. Examples of permissive/non-copyrighted audio include:
89
- - Public domain audio
90
- - Audio licensed under Apache, MIT, etc
91
- - Synthetic audio<sup>[1]</sup> generated by closed<sup>[2]</sup> TTS models from large providers<br/>
92
- [1] https://copyright.gov/ai/ai_policy_guidance.pdf<br/>
93
- [2] No synthetic audio from open TTS models or "custom voice clones"
94
 
95
- **Total Dataset Size:** A few hundred hours of audio
96
 
97
- **Total Training Cost:** About $1000 for 1000 hours of A100 80GB vRAM
98
 
99
- ### Creative Commons Attribution
100
 
101
- The following CC BY audio was part of the dataset used to train Kokoro v1.0.
102
 
103
- | Audio Data | Duration Used | License | Added to Training Set After |
104
- | ---------- | ------------- | ------- | --------------------------- |
105
- | [Koniwa](https://github.com/koniwa/koniwa) `tnc` | <1h | [CC BY 3.0](https://creativecommons.org/licenses/by/3.0/deed.ja) | v0.19 / 22 Nov 2024 |
106
- | [SIWIS](https://datashare.ed.ac.uk/handle/10283/2353) | <11h | [CC BY 4.0](https://datashare.ed.ac.uk/bitstream/handle/10283/2353/license_text) | v0.19 / 22 Nov 2024 |
107
 
108
- ### Acknowledgements
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
109
 
110
- - 🛠️ [@yl4579](https://huggingface.co/yl4579) for architecting StyleTTS 2.
111
- - 🏆 [@Pendrokar](https://huggingface.co/Pendrokar) for adding Kokoro as a contender in the TTS Spaces Arena.
112
- - 📊 Thank you to everyone who contributed synthetic training data.
113
- - ❤️ Special thanks to all compute sponsors.
114
- - 👾 Discord server: https://discord.gg/QuGxSWBfQy
115
- - 🪽 Kokoro is a Japanese word that translates to "heart" or "spirit". It is also the name of an [AI in the Terminator franchise](https://terminator.fandom.com/wiki/Kokoro).
116
 
117
- <img src="https://static0.gamerantimages.com/wordpress/wp-content/uploads/2024/08/terminator-zero-41-1.jpg" width="400" alt="kokoro" />
 
2
  license: apache-2.0
3
  language:
4
  - en
5
+ - zh
6
+ tags:
7
+ - video generation
8
+ - conversational video generation
9
+ - talking human video generation
10
+ pipeline_tag: image-to-video
11
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
12
 
13
+ <p align="center">
14
+ <img src="assets/logo2.jpeg" alt="MultiTalk" width="300"/>
15
+ </p>
16
+
17
+ # MeiGen-MultiTalk • Audio-Driven Multi-Person Conversational Video Generation
18
+
19
+
20
+
21
+ <p align="left">
22
+ <a href="https://meigen-ai.github.io/multi-talk/">
23
+ <img
24
+ src="https://img.shields.io/badge/MultiTalk-Website-0A66C2?logo=safari&logoColor=white" style="display: inline-block; vertical-align: middle;"
25
+ alt="MultiTalk Website"
26
+ />
27
+ </a>
28
+ <a href="https://arxiv.org/abs/2505.22647">
29
+ <img
30
+ src="https://img.shields.io/badge/MultiTalk-Paper-red?logo=arxiv&logoColor=red" style="display: inline-block; vertical-align: middle;"
31
+ alt="MultiTalk Paper on arXiv"
32
+ />
33
+ </a>
34
+ <a href="https://github.com/MeiGen-AI/MultiTalk" target="_blank" style="margin: 2px;">
35
+ <img
36
+ alt="Github" src="https://img.shields.io/badge/MultiTalk-Codebase-536af5?color=536af5&logo=github" style="display: inline-block; vertical-align: middle;"
37
+ alt="MultiTalk Codebase"
38
+ />
39
+ </a>
40
+
41
+ </p>
42
+
43
+ > We present **MultiTalk**, an open-source audio-driven multi-person conversational video generation model with the state-of-the-art lip synchronization accuracy.
44
+ > ​​Key features:​​
45
+ > - 💬 ​​Realistic Conversations​​ - Supports single & multi-person generation
46
+ > - 👥 ​​Interactive Character Control​​ - Direct virtual humans via prompts
47
+ > - 🎤 ​​Generalization Performances​​ - Supports the generation of cartoon character and singing
48
+ > - 📺 ​​Resolution Flexibility​​: 480p & 720p output at arbitrary aspect ratios
49
+ > - ⏱️ **Long Video Generation**: Support video generation up to 15 seconds
50
+
51
 
52
+ This repository hosts the model weights for **MultiTalk**. For installation, usage instructions, and further documentation, please visit our [GitHub repository](https://github.com/MeiGen-AI/MultiTalk).
 
 
 
53
 
 
54
 
 
55
 
 
56
 
 
57
 
 
58
 
59
+ ## Method
60
+ We propose a novel framework, MultiTalk, for audio-driven multi-person conversational video generation. We investigate several schemes for audio injection and introduce
61
+ the Label Rotary Position Embedding (L-RoPE) method. By assigning identical labels to audio embeddings and video latents, it effectively activates specific regions within the audio cross-attention
62
+ map, thereby resolving incorrect binding issues. To localize the region of the specified person, we introduce the adaptive person localization by computing the similarity
63
+ between the features of the given region of a person in the reference image and all the features of the whole video.
 
64
 
65
+ <p align="left"><img src="assets/pipe.png" width="80%"></p>
66
 
 
67
 
 
68
 
 
69
 
 
 
 
 
70
 
71
+ ## Citation
72
+ If you find our work helpful, please cite us.
73
+
74
+ ```
75
+ @article{kong2025let,
76
+ title={Let Them Talk: Audio-Driven Multi-Person Conversational Video Generation},
77
+ author={Kong, Zhe and Gao, Feng and Zhang, Yong and Kang, Zhuoliang and Wei, Xiaoming and Cai, Xunliang and Chen, Guanying and Luo, Wenhan},
78
+ journal={arXiv preprint arXiv:2505.22647},
79
+ year={2025}
80
+ }
81
+ ```
82
+
83
+
84
+
85
+ ## License Agreement
86
+ The models in this repository are licensed under the Apache 2.0 License. We claim no rights over the your generated contents, granting you the freedom to use them while ensuring that your usage complies with the provisions of this license. You are fully accountable for your use of the models, which must not involve sharing any content that violates applicable laws, causes harm to individuals or groups, disseminates personal information intended for harm, spreads misinformation, or targets vulnerable populations.
87
 
 
 
 
 
 
 
88
 
 
assets/logo.png ADDED

Git LFS Details

  • SHA256: 2fb97620f1515b94de007f5b5cde23e51aaa84a5cdc1eb91c021bb46b4cae3f0
  • Pointer size: 132 Bytes
  • Size of remote file: 3.31 MB
assets/logo2.jpeg ADDED

Git LFS Details

  • SHA256: 984efa12db10f378f37ba0576be90517658ed5c4a4146f2483121e9ae8fbd800
  • Pointer size: 131 Bytes
  • Size of remote file: 446 kB
assets/pipe.png ADDED

Git LFS Details

  • SHA256: dca19575d5c512b93d0eab2359cc75878da2064d4ef0e1f44aaf6accc04d6e0a
  • Pointer size: 132 Bytes
  • Size of remote file: 1.18 MB
diffusion_pytorch_model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
quant_models/dit_model_map_int8.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"text_embedding.0": {"weights": "qint8", "activations": "none"}, "text_embedding.2": {"weights": "qint8", "activations": "none"}, "blocks.0.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.0.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.0.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.0.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.0.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.0.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.0.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.0.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.0.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.0.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.0.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.0.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.0.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.0.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.0.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.1.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.1.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.1.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.1.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.1.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.1.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.1.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.1.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.1.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.1.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.1.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.1.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.1.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.1.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.1.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.2.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.2.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.2.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.2.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.2.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.2.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.2.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.2.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.2.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.2.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.2.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.2.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.2.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.2.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.2.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.3.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.3.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.3.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.3.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.3.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.3.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.3.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.3.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.3.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.3.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.3.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.3.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.3.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.3.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.3.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.4.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.4.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.4.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.4.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.4.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.4.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.4.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.4.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.4.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.4.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.4.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.4.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.4.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.4.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.4.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.5.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.5.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.5.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.5.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.5.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.5.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.5.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.5.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.5.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.5.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.5.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.5.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.5.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.5.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.5.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.6.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.6.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.6.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.6.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.6.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.6.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.6.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.6.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.6.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.6.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.6.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.6.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.6.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.6.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.6.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.7.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.7.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.7.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.7.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.7.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.7.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.7.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.7.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.7.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.7.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.7.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.7.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.7.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.7.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.7.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.8.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.8.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.8.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.8.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.8.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.8.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.8.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.8.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.8.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.8.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.8.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.8.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.8.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.8.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.8.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.9.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.9.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.9.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.9.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.9.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.9.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.9.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.9.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.9.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.9.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.9.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.9.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.9.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.9.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.9.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.10.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.10.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.10.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.10.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.10.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.10.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.10.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.10.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.10.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.10.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.10.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.10.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.10.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.10.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.10.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.11.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.11.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.11.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.11.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.11.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.11.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.11.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.11.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.11.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.11.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.11.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.11.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.11.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.11.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.11.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.12.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.12.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.12.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.12.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.12.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.12.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.12.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.12.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.12.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.12.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.12.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.12.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.12.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.12.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.12.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.13.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.13.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.13.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.13.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.13.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.13.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.13.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.13.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.13.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.13.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.13.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.13.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.13.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.13.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.13.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.14.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.14.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.14.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.14.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.14.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.14.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.14.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.14.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.14.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.14.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.14.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.14.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.14.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.14.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.14.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.15.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.15.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.15.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.15.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.15.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.15.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.15.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.15.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.15.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.15.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.15.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.15.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.15.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.15.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.15.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.16.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.16.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.16.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.16.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.16.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.16.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.16.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.16.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.16.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.16.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.16.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.16.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.16.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.16.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.16.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.17.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.17.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.17.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.17.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.17.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.17.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.17.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.17.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.17.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.17.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.17.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.17.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.17.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.17.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.17.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.18.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.18.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.18.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.18.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.18.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.18.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.18.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.18.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.18.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.18.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.18.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.18.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.18.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.18.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.18.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.19.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.19.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.19.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.19.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.19.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.19.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.19.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.19.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.19.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.19.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.19.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.19.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.19.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.19.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.19.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.20.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.20.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.20.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.20.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.20.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.20.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.20.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.20.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.20.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.20.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.20.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.20.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.20.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.20.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.20.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.21.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.21.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.21.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.21.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.21.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.21.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.21.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.21.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.21.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.21.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.21.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.21.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.21.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.21.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.21.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.22.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.22.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.22.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.22.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.22.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.22.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.22.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.22.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.22.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.22.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.22.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.22.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.22.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.22.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.22.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.23.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.23.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.23.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.23.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.23.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.23.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.23.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.23.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.23.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.23.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.23.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.23.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.23.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.23.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.23.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.24.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.24.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.24.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.24.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.24.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.24.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.24.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.24.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.24.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.24.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.24.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.24.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.24.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.24.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.24.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.25.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.25.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.25.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.25.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.25.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.25.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.25.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.25.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.25.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.25.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.25.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.25.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.25.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.25.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.25.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.26.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.26.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.26.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.26.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.26.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.26.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.26.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.26.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.26.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.26.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.26.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.26.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.26.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.26.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.26.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.27.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.27.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.27.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.27.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.27.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.27.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.27.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.27.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.27.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.27.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.27.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.27.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.27.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.27.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.27.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.28.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.28.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.28.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.28.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.28.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.28.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.28.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.28.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.28.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.28.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.28.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.28.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.28.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.28.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.28.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.29.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.29.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.29.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.29.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.29.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.29.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.29.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.29.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.29.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.29.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.29.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.29.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.29.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.29.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.29.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.30.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.30.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.30.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.30.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.30.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.30.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.30.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.30.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.30.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.30.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.30.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.30.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.30.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.30.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.30.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.31.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.31.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.31.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.31.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.31.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.31.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.31.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.31.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.31.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.31.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.31.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.31.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.31.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.31.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.31.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.32.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.32.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.32.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.32.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.32.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.32.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.32.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.32.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.32.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.32.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.32.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.32.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.32.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.32.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.32.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.33.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.33.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.33.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.33.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.33.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.33.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.33.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.33.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.33.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.33.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.33.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.33.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.33.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.33.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.33.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.34.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.34.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.34.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.34.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.34.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.34.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.34.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.34.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.34.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.34.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.34.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.34.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.34.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.34.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.34.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.35.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.35.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.35.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.35.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.35.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.35.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.35.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.35.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.35.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.35.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.35.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.35.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.35.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.35.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.35.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.36.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.36.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.36.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.36.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.36.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.36.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.36.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.36.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.36.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.36.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.36.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.36.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.36.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.36.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.36.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.37.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.37.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.37.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.37.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.37.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.37.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.37.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.37.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.37.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.37.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.37.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.37.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.37.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.37.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.37.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.38.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.38.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.38.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.38.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.38.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.38.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.38.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.38.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.38.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.38.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.38.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.38.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.38.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.38.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.38.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.39.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.39.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.39.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.39.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.39.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.39.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.39.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.39.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.39.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.39.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.39.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.39.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.39.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.39.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.39.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "audio_proj.proj1": {"weights": "qint8", "activations": "none"}, "audio_proj.proj1_vf": {"weights": "qint8", "activations": "none"}, "audio_proj.proj2": {"weights": "qint8", "activations": "none"}, "audio_proj.proj3": {"weights": "qint8", "activations": "none"}}
quant_models/quantization_map_fp8_FusionX.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"text_embedding.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "text_embedding.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.24.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.25.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.26.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.27.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.28.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.29.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.30.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.31.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.32.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.33.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.34.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.35.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.36.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.37.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.38.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.self_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.self_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.self_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.self_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.cross_attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.cross_attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.cross_attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.cross_attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.cross_attn.k_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.cross_attn.v_img": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.ffn.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.ffn.2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.audio_cross_attn.q_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.audio_cross_attn.proj": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.39.audio_cross_attn.kv_linear": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "audio_proj.proj1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "audio_proj.proj1_vf": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "audio_proj.proj2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "audio_proj.proj3": {"weights": "qfloat8_e4m3fn", "activations": "none"}}
quant_models/quantization_map_int8_FusionX.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"text_embedding.0": {"weights": "qint8", "activations": "none"}, "text_embedding.2": {"weights": "qint8", "activations": "none"}, "blocks.0.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.0.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.0.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.0.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.0.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.0.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.0.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.0.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.0.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.0.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.0.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.0.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.0.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.0.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.0.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.1.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.1.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.1.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.1.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.1.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.1.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.1.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.1.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.1.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.1.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.1.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.1.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.1.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.1.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.1.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.2.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.2.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.2.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.2.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.2.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.2.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.2.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.2.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.2.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.2.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.2.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.2.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.2.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.2.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.2.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.3.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.3.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.3.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.3.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.3.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.3.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.3.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.3.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.3.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.3.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.3.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.3.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.3.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.3.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.3.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.4.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.4.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.4.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.4.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.4.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.4.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.4.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.4.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.4.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.4.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.4.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.4.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.4.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.4.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.4.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.5.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.5.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.5.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.5.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.5.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.5.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.5.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.5.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.5.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.5.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.5.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.5.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.5.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.5.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.5.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.6.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.6.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.6.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.6.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.6.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.6.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.6.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.6.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.6.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.6.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.6.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.6.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.6.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.6.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.6.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.7.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.7.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.7.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.7.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.7.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.7.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.7.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.7.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.7.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.7.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.7.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.7.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.7.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.7.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.7.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.8.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.8.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.8.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.8.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.8.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.8.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.8.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.8.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.8.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.8.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.8.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.8.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.8.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.8.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.8.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.9.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.9.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.9.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.9.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.9.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.9.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.9.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.9.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.9.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.9.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.9.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.9.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.9.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.9.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.9.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.10.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.10.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.10.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.10.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.10.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.10.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.10.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.10.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.10.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.10.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.10.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.10.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.10.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.10.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.10.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.11.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.11.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.11.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.11.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.11.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.11.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.11.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.11.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.11.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.11.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.11.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.11.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.11.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.11.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.11.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.12.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.12.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.12.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.12.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.12.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.12.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.12.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.12.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.12.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.12.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.12.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.12.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.12.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.12.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.12.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.13.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.13.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.13.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.13.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.13.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.13.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.13.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.13.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.13.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.13.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.13.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.13.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.13.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.13.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.13.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.14.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.14.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.14.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.14.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.14.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.14.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.14.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.14.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.14.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.14.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.14.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.14.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.14.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.14.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.14.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.15.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.15.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.15.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.15.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.15.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.15.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.15.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.15.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.15.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.15.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.15.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.15.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.15.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.15.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.15.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.16.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.16.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.16.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.16.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.16.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.16.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.16.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.16.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.16.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.16.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.16.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.16.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.16.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.16.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.16.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.17.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.17.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.17.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.17.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.17.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.17.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.17.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.17.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.17.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.17.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.17.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.17.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.17.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.17.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.17.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.18.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.18.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.18.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.18.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.18.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.18.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.18.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.18.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.18.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.18.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.18.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.18.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.18.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.18.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.18.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.19.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.19.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.19.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.19.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.19.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.19.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.19.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.19.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.19.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.19.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.19.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.19.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.19.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.19.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.19.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.20.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.20.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.20.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.20.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.20.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.20.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.20.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.20.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.20.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.20.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.20.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.20.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.20.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.20.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.20.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.21.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.21.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.21.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.21.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.21.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.21.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.21.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.21.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.21.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.21.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.21.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.21.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.21.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.21.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.21.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.22.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.22.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.22.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.22.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.22.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.22.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.22.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.22.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.22.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.22.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.22.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.22.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.22.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.22.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.22.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.23.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.23.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.23.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.23.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.23.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.23.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.23.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.23.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.23.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.23.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.23.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.23.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.23.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.23.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.23.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.24.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.24.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.24.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.24.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.24.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.24.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.24.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.24.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.24.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.24.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.24.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.24.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.24.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.24.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.24.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.25.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.25.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.25.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.25.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.25.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.25.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.25.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.25.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.25.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.25.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.25.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.25.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.25.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.25.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.25.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.26.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.26.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.26.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.26.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.26.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.26.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.26.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.26.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.26.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.26.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.26.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.26.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.26.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.26.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.26.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.27.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.27.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.27.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.27.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.27.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.27.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.27.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.27.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.27.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.27.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.27.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.27.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.27.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.27.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.27.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.28.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.28.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.28.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.28.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.28.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.28.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.28.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.28.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.28.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.28.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.28.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.28.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.28.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.28.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.28.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.29.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.29.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.29.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.29.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.29.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.29.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.29.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.29.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.29.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.29.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.29.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.29.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.29.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.29.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.29.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.30.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.30.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.30.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.30.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.30.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.30.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.30.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.30.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.30.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.30.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.30.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.30.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.30.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.30.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.30.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.31.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.31.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.31.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.31.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.31.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.31.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.31.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.31.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.31.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.31.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.31.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.31.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.31.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.31.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.31.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.32.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.32.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.32.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.32.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.32.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.32.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.32.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.32.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.32.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.32.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.32.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.32.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.32.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.32.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.32.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.33.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.33.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.33.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.33.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.33.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.33.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.33.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.33.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.33.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.33.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.33.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.33.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.33.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.33.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.33.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.34.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.34.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.34.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.34.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.34.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.34.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.34.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.34.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.34.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.34.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.34.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.34.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.34.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.34.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.34.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.35.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.35.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.35.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.35.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.35.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.35.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.35.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.35.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.35.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.35.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.35.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.35.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.35.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.35.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.35.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.36.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.36.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.36.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.36.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.36.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.36.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.36.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.36.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.36.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.36.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.36.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.36.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.36.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.36.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.36.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.37.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.37.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.37.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.37.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.37.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.37.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.37.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.37.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.37.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.37.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.37.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.37.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.37.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.37.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.37.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.38.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.38.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.38.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.38.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.38.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.38.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.38.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.38.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.38.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.38.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.38.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.38.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.38.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.38.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.38.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "blocks.39.self_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.39.self_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.39.self_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.39.self_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.39.cross_attn.q": {"weights": "qint8", "activations": "none"}, "blocks.39.cross_attn.k": {"weights": "qint8", "activations": "none"}, "blocks.39.cross_attn.v": {"weights": "qint8", "activations": "none"}, "blocks.39.cross_attn.o": {"weights": "qint8", "activations": "none"}, "blocks.39.cross_attn.k_img": {"weights": "qint8", "activations": "none"}, "blocks.39.cross_attn.v_img": {"weights": "qint8", "activations": "none"}, "blocks.39.ffn.0": {"weights": "qint8", "activations": "none"}, "blocks.39.ffn.2": {"weights": "qint8", "activations": "none"}, "blocks.39.audio_cross_attn.q_linear": {"weights": "qint8", "activations": "none"}, "blocks.39.audio_cross_attn.proj": {"weights": "qint8", "activations": "none"}, "blocks.39.audio_cross_attn.kv_linear": {"weights": "qint8", "activations": "none"}, "audio_proj.proj1": {"weights": "qint8", "activations": "none"}, "audio_proj.proj1_vf": {"weights": "qint8", "activations": "none"}, "audio_proj.proj2": {"weights": "qint8", "activations": "none"}, "audio_proj.proj3": {"weights": "qint8", "activations": "none"}}
quant_models/t5_map_fp8.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"blocks.0.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.0.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.1.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.2.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.3.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.4.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.5.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.6.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.7.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.8.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.9.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.10.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.11.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.12.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.13.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.14.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.15.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.16.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.17.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.18.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.19.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.20.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.21.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.22.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.attn.q": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.attn.k": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.attn.v": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.attn.o": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.ffn.gate.0": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.ffn.fc1": {"weights": "qfloat8_e4m3fn", "activations": "none"}, "blocks.23.ffn.fc2": {"weights": "qfloat8_e4m3fn", "activations": "none"}}
quant_models/t5_map_int8.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"blocks.0.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.0.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.0.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.0.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.0.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.0.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.0.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.1.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.1.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.1.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.1.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.1.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.1.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.1.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.2.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.2.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.2.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.2.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.2.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.2.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.2.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.3.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.3.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.3.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.3.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.3.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.3.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.3.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.4.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.4.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.4.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.4.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.4.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.4.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.4.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.5.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.5.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.5.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.5.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.5.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.5.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.5.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.6.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.6.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.6.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.6.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.6.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.6.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.6.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.7.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.7.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.7.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.7.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.7.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.7.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.7.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.8.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.8.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.8.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.8.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.8.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.8.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.8.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.9.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.9.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.9.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.9.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.9.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.9.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.9.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.10.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.10.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.10.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.10.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.10.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.10.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.10.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.11.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.11.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.11.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.11.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.11.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.11.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.11.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.12.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.12.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.12.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.12.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.12.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.12.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.12.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.13.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.13.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.13.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.13.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.13.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.13.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.13.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.14.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.14.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.14.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.14.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.14.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.14.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.14.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.15.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.15.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.15.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.15.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.15.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.15.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.15.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.16.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.16.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.16.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.16.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.16.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.16.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.16.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.17.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.17.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.17.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.17.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.17.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.17.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.17.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.18.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.18.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.18.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.18.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.18.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.18.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.18.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.19.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.19.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.19.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.19.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.19.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.19.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.19.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.20.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.20.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.20.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.20.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.20.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.20.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.20.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.21.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.21.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.21.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.21.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.21.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.21.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.21.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.22.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.22.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.22.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.22.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.22.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.22.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.22.ffn.fc2": {"weights": "qint8", "activations": "none"}, "blocks.23.attn.q": {"weights": "qint8", "activations": "none"}, "blocks.23.attn.k": {"weights": "qint8", "activations": "none"}, "blocks.23.attn.v": {"weights": "qint8", "activations": "none"}, "blocks.23.attn.o": {"weights": "qint8", "activations": "none"}, "blocks.23.ffn.gate.0": {"weights": "qint8", "activations": "none"}, "blocks.23.ffn.fc1": {"weights": "qint8", "activations": "none"}, "blocks.23.ffn.fc2": {"weights": "qint8", "activations": "none"}}