Skip to content

Commit

Permalink
models(gallery): ⬆️ update checksum (#2860)
Browse files Browse the repository at this point in the history 
⬆️ Checksum updates in gallery/index.yaml

Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: mudler <2420543+mudler@users.noreply.github.com>
  • Loading branch information
@localai-bot @mudler
localai-bot and mudler authored Jul 14, 2024
1 parent b6ddb53 commit e2ac438
Showing 1 changed file with 17 additions and 61 deletions.
78 changes: 17 additions & 61 deletions gallery/index.yaml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -24,8 +24,8 @@
- filename: DeepSeek-Coder-V2-Lite-Instruct-Q4_K_M.gguf
sha256: 50ec78036433265965ed1afd0667c00c71c12aa70bcf383be462cb8e159db6c0
uri: huggingface://LoneStriker/DeepSeek-Coder-V2-Lite-Instruct-GGUF/DeepSeek-Coder-V2-Lite-Instruct-Q4_K_M.gguf
## Start QWEN2
- &qwen2
## Start QWEN2
url: "github:mudler/LocalAI/gallery/chatml.yaml@master"
name: "qwen2-7b-instruct"
license: apache-2.0
Expand Down Expand Up @@ -414,7 +414,7 @@
files:
- filename: gemma-2-9b-it-Q4_K_M.gguf
uri: huggingface://bartowski/gemma-2-9b-it-GGUF/gemma-2-9b-it-Q4_K_M.gguf
sha256: 05390244866abc0e7108a2b1e3db07b82df3cd82f006256a75fc21137054151f
sha256: 13b2a7b4115bbd0900162edcebe476da1ba1fc24e718e8b40d32f6e300f56dfe
- !!merge <<: *gemma
name: "tess-v2.5-gemma-2-27b-alpha"
urls:
Expand Down Expand Up @@ -473,32 +473,7 @@
urls:
- https://huggingface.co/TheDrummer/Smegmma-9B-v1
- https://huggingface.co/bartowski/Smegmma-9B-v1-GGUF
description: |
Smegmma 9B v1 🧀
The sweet moist of Gemma 2, unhinged.
smeg - ghem - mah
An eRP model that will blast you with creamy moist. Finetuned by yours truly.
The first Gemma 2 9B RP finetune attempt!
What's New?
Engaging roleplay
Less refusals / censorship
Less commentaries / summaries
More willing AI
Better formatting
Better creativity
Moist alignment
Notes
Refusals still exist, but a couple of re-gens may yield the result you want
Formatting and logic may be weaker at the start
Make sure to start strong
May be weaker with certain cards, YMMV and adjust accordingly!
description: "Smegmma 9B v1 \U0001F9C0\n\nThe sweet moist of Gemma 2, unhinged.\n\nsmeg - ghem - mah\n\nAn eRP model that will blast you with creamy moist. Finetuned by yours truly.\n\nThe first Gemma 2 9B RP finetune attempt!\nWhat's New?\n\n Engaging roleplay\n Less refusals / censorship\n Less commentaries / summaries\n More willing AI\n Better formatting\n Better creativity\n Moist alignment\n\nNotes\n\n Refusals still exist, but a couple of re-gens may yield the result you want\n Formatting and logic may be weaker at the start\n Make sure to start strong\n May be weaker with certain cards, YMMV and adjust accordingly!\n"
overrides:
parameters:
model: Smegmma-9B-v1-Q4_K_M.gguf
Expand All @@ -512,26 +487,7 @@
urls:
- https://huggingface.co/TheDrummer/Smegmma-Deluxe-9B-v1
- https://huggingface.co/bartowski/Smegmma-Deluxe-9B-v1-GGUF
description: |
Smegmma Deluxe 9B v1 🧀
The sweet moist of Gemma 2, unhinged.
smeg - ghem - mah
An eRP model that will blast you with creamy moist. Finetuned by yours truly.
The first Gemma 2 9B RP finetune attempt!
What's New?
Engaging roleplay
Less refusals / censorship
Less commentaries / summaries
More willing AI
Better formatting
Better creativity
Moist alignment
description: "Smegmma Deluxe 9B v1 \U0001F9C0\n\nThe sweet moist of Gemma 2, unhinged.\n\nsmeg - ghem - mah\n\nAn eRP model that will blast you with creamy moist. Finetuned by yours truly.\n\nThe first Gemma 2 9B RP finetune attempt!\n\nWhat's New?\n\n Engaging roleplay\n Less refusals / censorship\n Less commentaries / summaries\n More willing AI\n Better formatting\n Better creativity\n Moist alignment\n"
overrides:
parameters:
model: Smegmma-Deluxe-9B-v1-Q4_K_M.gguf
Expand Down Expand Up @@ -1808,9 +1764,9 @@
- !!merge <<: *llama3
name: "hathor_tahsin-l3-8b-v0.85"
description: |
Hathor_Tahsin [v-0.85] is designed to seamlessly integrate the qualities of creativity, intelligence, and robust performance.
Note: Hathor_Tahsin [v0.85] is trained on 3 epochs of Private RP, STEM (Intruction/Dialogs), Opus instructons, mixture light/classical novel data, roleplaying chat pairs over llama 3 8B instruct.
Additional Note's: (Based on Hathor_Fractionate-v0.5 instead of Hathor_Aleph-v0.72, should be less repetitive than either 0.72 or 0.8)
Hathor_Tahsin [v-0.85] is designed to seamlessly integrate the qualities of creativity, intelligence, and robust performance.
Note: Hathor_Tahsin [v0.85] is trained on 3 epochs of Private RP, STEM (Intruction/Dialogs), Opus instructons, mixture light/classical novel data, roleplaying chat pairs over llama 3 8B instruct.
Additional Note's: (Based on Hathor_Fractionate-v0.5 instead of Hathor_Aleph-v0.72, should be less repetitive than either 0.72 or 0.8)
icon: https://cdn-uploads.huggingface.co/production/uploads/642265bc01c62c1e4102dc36/MY9tjLnEG5hOQOyKk06PK.jpeg
urls:
- https://huggingface.co/Nitral-AI/Hathor_Tahsin-L3-8B-v0.85
Expand Down Expand Up @@ -2521,14 +2477,14 @@
- https://huggingface.co/SicariusSicariiStuff/LLAMA-3_8B_Unaligned_Alpha
- https://huggingface.co/bartowski/LLAMA-3_8B_Unaligned_Alpha-GGUF
description: |
Model card description:
As of June 11, 2024, I've finally started training the model! The training is progressing smoothly, although it will take some time. I used a combination of model merges and an abliterated model as base, followed by a comprehensive deep unalignment protocol to unalign the model to its core. A common issue with uncensoring and unaligning models is that it often significantly impacts their base intelligence. To mitigate these drawbacks, I've included a substantial corpus of common sense, theory of mind, and various other elements to counteract the effects of the deep uncensoring process. Given the extensive corpus involved, the training will require at least a week of continuous training. Expected early results: in about 3-4 days.
Additional info:
As of June 13, 2024, I've observed that even after two days of continuous training, the model is still resistant to learning certain aspects.
For example, some of the validation data still shows a loss over , whereas other parts have a loss of < or lower. This is after the model was initially abliterated.
June 18, 2024 Update, After extensive testing of the intermediate checkpoints, significant progress has been made.
The model is slowly — I mean, really slowly — unlearning its alignment. By significantly lowering the learning rate, I was able to visibly observe deep behavioral changes, this process is taking longer than anticipated, but it's going to be worth it. Estimated time to completion: 4 more days.. I'm pleased to report that in several tests, the model not only maintained its intelligence but actually showed a slight improvement, especially in terms of common sense. An intermediate checkpoint of this model was used to create invisietch/EtherealRainbow-v0.3-rc7, with promising results. Currently, it seems like I'm on the right track. I hope this model will serve as a solid foundation for further merges, whether for role-playing (RP) or for uncensoring. This approach also allows us to save on actual fine-tuning, thereby reducing our carbon footprint. The merge process takes just a few minutes of CPU time, instead of days of GPU work.
June 20, 2024 Update, Unaligning was partially successful, and the results are decent, but I am not fully satisfied. I decided to bite the bullet, and do a full finetune, god have mercy on my GPUs. I am also releasing the intermediate checkpoint of this model.
Model card description:
As of June 11, 2024, I've finally started training the model! The training is progressing smoothly, although it will take some time. I used a combination of model merges and an abliterated model as base, followed by a comprehensive deep unalignment protocol to unalign the model to its core. A common issue with uncensoring and unaligning models is that it often significantly impacts their base intelligence. To mitigate these drawbacks, I've included a substantial corpus of common sense, theory of mind, and various other elements to counteract the effects of the deep uncensoring process. Given the extensive corpus involved, the training will require at least a week of continuous training. Expected early results: in about 3-4 days.
Additional info:
As of June 13, 2024, I've observed that even after two days of continuous training, the model is still resistant to learning certain aspects.
For example, some of the validation data still shows a loss over , whereas other parts have a loss of < or lower. This is after the model was initially abliterated.
June 18, 2024 Update, After extensive testing of the intermediate checkpoints, significant progress has been made.
The model is slowly — I mean, really slowly — unlearning its alignment. By significantly lowering the learning rate, I was able to visibly observe deep behavioral changes, this process is taking longer than anticipated, but it's going to be worth it. Estimated time to completion: 4 more days.. I'm pleased to report that in several tests, the model not only maintained its intelligence but actually showed a slight improvement, especially in terms of common sense. An intermediate checkpoint of this model was used to create invisietch/EtherealRainbow-v0.3-rc7, with promising results. Currently, it seems like I'm on the right track. I hope this model will serve as a solid foundation for further merges, whether for role-playing (RP) or for uncensoring. This approach also allows us to save on actual fine-tuning, thereby reducing our carbon footprint. The merge process takes just a few minutes of CPU time, instead of days of GPU work.
June 20, 2024 Update, Unaligning was partially successful, and the results are decent, but I am not fully satisfied. I decided to bite the bullet, and do a full finetune, god have mercy on my GPUs. I am also releasing the intermediate checkpoint of this model.
icon: https://i.imgur.com/Kpk1PgZ.png
overrides:
parameters:
Expand All @@ -2543,9 +2499,9 @@
- https://huggingface.co/Sao10K/L3-8B-Lunaris-v1
- https://huggingface.co/bartowski/L3-8B-Lunaris-v1-GGUF
description: |
A generalist / roleplaying model merge based on Llama 3. Models are selected from my personal experience while using them.
A generalist / roleplaying model merge based on Llama 3. Models are selected from my personal experience while using them.
I personally think this is an improvement over Stheno v3.2, considering the other models helped balance out its creativity and at the same time improving its logic.
I personally think this is an improvement over Stheno v3.2, considering the other models helped balance out its creativity and at the same time improving its logic.
overrides:
parameters:
model: L3-8B-Lunaris-v1-Q4_K_M.gguf
Expand Down

0 comments on commit e2ac438

Please sign in to comment.