Update README.md
Browse files
README.md
CHANGED
|
@@ -30,24 +30,22 @@ We'll release multiple HumanPet models:
|
|
| 30 |
| 7 | HumanPet Tool 1.7B | Less robotic, full visual emotion, full personality, tools | Stage 6 + **translated tool datasets (stage 5)** | Minimal | Low effort. Use the translator (stage 5) to create a silly tool dataset. *It is not sure if this stage will be made. It all depends on the results of stage 5 and 6.* | 2027 |
|
| 31 |
| 8 | HumanPet2 Instruct *???* B | Less robotic, full visual emotion, full personality | Stage 6 + **other datasets translated (stage 5)** | Close to none | Low effort. The final model (stage 6) reinforced with datasets that contain less hallucination, translated with the translator (stage 5). *It is not sure if this stage will be made. We might use another method. It all depends on the results of stage 5 and 6.* | 2027 - early 2028 |
|
| 32 |
|
| 33 |
-
|
| 34 |
-
Each stage depends on the last or recent stages, and as such, if one stage is very disappointing (highly likely, considering LLMs fail to deliver well most of the time), the others might never continue. Though, we'll try to prevent that.
|
| 35 |
The **"X1"/"X2"/etc.** stands for **"experimental model stage 1"/"experimental model stage 2"** etc.
|
| 36 |
|
| 37 |
-
|
| 38 |
|
| 39 |
-
The reason why we include a translator at stage 3 and 5 mainly has to do with being able to turn other datasets into silly ones. This means that we
|
| 40 |
-
This stage is uncertain though, and may not work that well. If it doesn't, stage 7 and 6 will likely be dropped too.
|
| 41 |
|
| 42 |
-
The
|
| 43 |
|
| 44 |
| Stage | Name | Time |
|
| 45 |
| ----- | ---- | ---- |
|
| 46 |
| 1 | HumanPet X1 1.7B | Now |
|
| 47 |
-
| 2 |
|
| 48 |
| 3 | HumanPet Translator Semi 1.7B | March-April 2026 |
|
| 49 |
-
| 4 | HumanPet X3 1.7B | 2026 |
|
| 50 |
-
| 6 | HumanPet Instruct 1.7B | 2027 |
|
| 51 |
|
| 52 |
## Progress
|
| 53 |
|
|
@@ -65,16 +63,16 @@ Progress per asset:
|
|
| 65 |
| Model | Re-train: HumanPet X2.2 1.7B | π’ Done |
|
| 66 |
| Dataset | HumanPet Translator Semi 4.3k | π’ Done |
|
| 67 |
| Model | HumanPet Translator Semi 1.7B | π΄ Done, mistakes found |
|
| 68 |
-
| Dataset | HumanPet X3 *???* k |
|
| 69 |
-
| Model | HumanPet X3 1.7B |
|
| 70 |
-
| Dataset | HumanPet Translator *???* k |
|
| 71 |
-
| Model | HumanPet Translator 1.7B |
|
| 72 |
-
| Dataset | HumanPet Instruct *???* k |
|
| 73 |
-
| Model | HumanPet Instruct 1.7B |
|
| 74 |
-
| Dataset | HumanPet Tool *???* k |
|
| 75 |
-
| Model | HumanPet Tool 1.7B |
|
| 76 |
-
| Dataset | HumanPet2 Instruct *???* k |
|
| 77 |
-
| Model | HumanPet2 Instruct *???* B |
|
| 78 |
|
| 79 |
*(31% done.)*
|
| 80 |
|
|
|
|
| 30 |
| 7 | HumanPet Tool 1.7B | Less robotic, full visual emotion, full personality, tools | Stage 6 + **translated tool datasets (stage 5)** | Minimal | Low effort. Use the translator (stage 5) to create a silly tool dataset. *It is not sure if this stage will be made. It all depends on the results of stage 5 and 6.* | 2027 |
|
| 31 |
| 8 | HumanPet2 Instruct *???* B | Less robotic, full visual emotion, full personality | Stage 6 + **other datasets translated (stage 5)** | Close to none | Low effort. The final model (stage 6) reinforced with datasets that contain less hallucination, translated with the translator (stage 5). *It is not sure if this stage will be made. We might use another method. It all depends on the results of stage 5 and 6.* | 2027 - early 2028 |
|
| 32 |
|
| 33 |
+
Unfortunately, this project **has come to an end**. Note that **all stages are experimental.** Stage 1, 2 and 3 are done, but all the other stages **will most likely never complete.**
|
|
|
|
| 34 |
The **"X1"/"X2"/etc.** stands for **"experimental model stage 1"/"experimental model stage 2"** etc.
|
| 35 |
|
| 36 |
+
Because this is the end of the HumanPet experiment, we're releasing all model files soon.
|
| 37 |
|
| 38 |
+
The reason why we wanted to include a translator at stage 3 and 5 mainly has to do with being able to turn other datasets into silly ones. This means that we could've used tool, chain-of-thought or other instruct datasets while keeping the personality in tact.
|
|
|
|
| 39 |
|
| 40 |
+
The roadmap was:
|
| 41 |
|
| 42 |
| Stage | Name | Time |
|
| 43 |
| ----- | ---- | ---- |
|
| 44 |
| 1 | HumanPet X1 1.7B | Now |
|
| 45 |
+
| 2 | HumanPet X2 1.7B | March 2026 |
|
| 46 |
| 3 | HumanPet Translator Semi 1.7B | March-April 2026 |
|
| 47 |
+
| 4 | HumanPet X3 1.7B *(canceled)* | 2026 |
|
| 48 |
+
| 6 | HumanPet Instruct 1.7B *(canceled)* | 2027 |
|
| 49 |
|
| 50 |
## Progress
|
| 51 |
|
|
|
|
| 63 |
| Model | Re-train: HumanPet X2.2 1.7B | π’ Done |
|
| 64 |
| Dataset | HumanPet Translator Semi 4.3k | π’ Done |
|
| 65 |
| Model | HumanPet Translator Semi 1.7B | π΄ Done, mistakes found |
|
| 66 |
+
| Dataset | HumanPet X3 *???* k | π΄ Canceled |
|
| 67 |
+
| Model | HumanPet X3 1.7B | π΄ Canceled |
|
| 68 |
+
| Dataset | HumanPet Translator *???* k | π΄ Canceled |
|
| 69 |
+
| Model | HumanPet Translator 1.7B | π΄ Canceled |
|
| 70 |
+
| Dataset | HumanPet Instruct *???* k | π΄ Canceled |
|
| 71 |
+
| Model | HumanPet Instruct 1.7B | π΄ Canceled |
|
| 72 |
+
| Dataset | HumanPet Tool *???* k | π΄ Canceled |
|
| 73 |
+
| Model | HumanPet Tool 1.7B | π΄ Canceled |
|
| 74 |
+
| Dataset | HumanPet2 Instruct *???* k | π΄ Canceled |
|
| 75 |
+
| Model | HumanPet2 Instruct *???* B | π΄ Canceled |
|
| 76 |
|
| 77 |
*(31% done.)*
|
| 78 |
|