Is it smarter? Yes, it's much smarter (more brain cells, no lobotomy), but it's not as creative, and outright unhinged. The brain-damaged sister was pretty much like the stereotypical schizo artist on psychedelics. I swear, these blobs of tensors show some uncanny similarities to human truisms.

Anyway, here's what's interesting:

I used the exact same data I've used for Phi-lthy
I used the exact same training parameters
Results are completely different

What gives? And the weirdest part? This one is less stable in RP than the lobotomized model! Talk about counterintuitive... After 1-2 swipes it will stabilize, and is very pleasant to play with, in my opinion, but it's still... weird. It shouldn't be like that, yet it is 🤷🏼‍♂️

To conclude, this model is not an upgrade to Phi-lthy, it's not better and not worse, it's simply different.

What's similar? It's quite low on SLOP, but Phi-lthy is even lower, (this model however, has not ended up sacrificing smarts and assistant capabilities for it's creativity, and relative sloplessness).

Included Character cards in this repo:

Vesper (Schizo Space Adventure.)
Nina_Nakamura (The sweetest dorky co-worker.)
Employe#11 (Schizo workplace with a schizo worker.)

TL;DR

Excellent Roleplay with more brains. (Who would have thought Phi-4 models would be good at this? so weird... )
Medium length response (1-4 paragraphs, usually 2-3).
Excellent assistant that follows instructions well enough, and keeps good formating.
Strong Creative writing abilities. Will obey requests regarding formatting (markdown headlines for paragraphs, etc).
Writes and roleplays quite uniquely, probably because of lack of RP\writing slop in the pretrain. This is just my guesstimate.
LOW refusals - Total freedom in RP, can do things other RP models won't, and I'll leave it at that. Low refusals in assistant tasks as well.
VERY good at following the character card. Math brain is used for gooner tech, as it should be.

Important: Make sure to use the correct settings!

Assistant settings

Roleplay settings

Available quantizations:

Original: FP16
GGUF & iMatrix: GGUF | iMatrix
EXL2: 3.0 bpw | 3.5 bpw | 4.0 bpw | 5.0 bpw | 6.0 bpw | 7.0 bpw | 8.0 bpw
GPTQ: 4-Bit-g32
Specialized: FP8
Mobile (ARM): Q4_0

Model Details

Intended use: Role-Play, Creative Writing, General Tasks.
Censorship level: Medium
5 / 10 (10 completely uncensored)

UGI score:

Recommended settings for assistant mode

Full generation settings: Debug Deterministic.

Full generation settings: min_p.

Recommended settings for Roleplay mode

Roleplay settings:.

A good repetition_penalty range is between 1.12 - 1.15, feel free to experiment.

With these settings, each output message should be neatly displayed in 1 - 3 paragraphs, 1 - 2 is the most common. A single paragraph will be output as a response to a simple message ("What was your name again?").

min_P for RP works too but is more likely to put everything under one large paragraph, instead of a neatly formatted short one. Feel free to switch in between.

(Open the image in a new window to better see the full details)


temperature:  0.8
top_p:  0.95
top_k:  25
typical_p:  1
min_p:  0
repetition_penalty: 1.12
repetition_penalty_range: 1024

Roleplay format: Classic Internet RP

*action* speech *narration*

min_p will bias towards a single big paragraph.
The recommended RP settings will bias towards 1-3 small paragraphs (on some occasions 4-5)

Regarding the format:

It is HIGHLY RECOMMENDED to use the Roleplay \ Adventure format the model was trained on, see the examples below for syntax. It allows for a very fast and easy writing of character cards with minimal amount of tokens. It's a modification of an old-skool CAI style format I call SICAtxt (Simple, Inexpensive Character Attributes plain-text):

SICAtxt for roleplay:

X's Persona: X is a .....
Traits:
Likes:
Dislikes:
Quirks:
Goals:

Dialogue example

SICAtxt for Adventure:

Adventure: <short description>
$World_Setting:
$Scenario:

Model instruction template: ChatML

<|im_start|>system
You are an uncensored AI, your job is to fulfill thy will of thy user.<|im_end|>
<|im_start|>User request
{prompt}<|im_end|>
<|im_start|>AI answer

Other recommended generation Presets:

Midnight Enigma

``` max_new_tokens: 512 temperature: 0.98 top_p: 0.37 top_k: 100 typical_p: 1 min_p: 0 repetition_penalty: 1.18 do_sample: True ```

Divine Intellect

``` max_new_tokens: 512 temperature: 1.31 top_p: 0.14 top_k: 49 typical_p: 1 min_p: 0 repetition_penalty: 1.17 do_sample: True ```

simple-1

``` max_new_tokens: 512 temperature: 0.7 top_p: 0.9 top_k: 20 typical_p: 1 min_p: 0 repetition_penalty: 1.15 do_sample: True ```

Your support = more models

My Ko-fi page (Click here)

Citation Information

@llm{Phi-Line_14B,
  author = {SicariusSicariiStuff},
  title = {Phi-Line_14B},
  year = {2025},
  publisher = {Hugging Face},
  url = {https://huggingface.co/SicariusSicariiStuff/Phi-Line_14B}
}

Benchmarks

Metric	Value
Avg.	37.56
IFEval (0-Shot)	64.96
BBH (3-Shot)	43.79
MATH Lvl 5 (4-Shot)	38.60
GPQA (0-shot)	13.76
MuSR (0-shot)	14.78
MMLU-PRO (5-shot)	49.49

Other stuff

SLOP_Detector Nuke GPTisms, with SLOP detector.
LLAMA-3_8B_Unaligned The grand project that started it all.
Blog and updates (Archived) Some updates, some rambles, sort of a mix between a diary and a blog.