L3-Grand-STORY-16.5B Ultra Quality - A triple model, triple step merge at Full Precision F32. Imatrix Neo Class 2.
WARNING: NSFW. Ultra Detailed. Graphic HORROR, VIOLENCE. Extreme swearing. UNCENSORED. SMART.
I took the original models in "L3-Stheno-Maid-Blackroot 8B" and completely rebuilt it a new pass-through merge (everything preserved)
and blew it out to over 16.5 billion parameters - 642 tensors, 71 layers (8B original has 32 layers) at full float 32 precision.
Using these three models I build TWO custom Llama3 models:
Grand Horror 16.5B (
here ) and
Grand Story 16.5B ALPHA (unreleased) then merged these together with a "smoothing step" captured at F32 precision.
(formula below, along with critical merge model notes and theory)
The result is a model that is far more stable, far more capable than any of the 3 models originally and it is
greater than the "sum" of two 16.5B models mentioned. Compared to Grand Horror 16.5B is it over 25000 points
lower (IQ4XS) in perplexity (lower is better) or 2.5 full levels of magnitude lower.
It is tougher, stronger and can handle a far wider range of operating conditions - from temp .1 to temp 5 all day long.
The F32 precision (along with full F32 transfer to the ggufs) increases the performance even further.
This added precision increases the model's depth and nuance including "world" perception, real time in the moment
similes and metaphors, description of the 5 senses, and word choice in general.
The model's grasp of "facts" and where to use them have also improved and likewise "facts" it makes up are far more "believable".
Sentence structure and variety are also significantly improved, as are paragraph structure and variety too.
The result is a take no prisoners, totally uncensored, fiction writing monster and roleplay master as well
just about... any general fiction activity "AI guru" including scene generation and scene continuation.
This model is capable of horror, science fiction, romance - you name it.
This model has a very strong VIDIDINESS bias. It generates extremely vivid prose, description, and dialog as well
as in the moment metaphors and similes and it rarely uses "cliches".
It also has a STRONG horror bias, although it will generate content for almost any genre. That being said
if there is a "hint" of things going wrong... they will.
It will also swear (R-18) like there is no tomorrow at times and "dark" characters will be VERY dark so to speak.
Model excels in details (real and "constructed"), descriptions, similes and metaphors including dates, times
and "fictional history" that sounds "real".
I would also say it can have a sense of humor ... ah... dark humor.
With all this being said, this model has an uncanny sense of "there" , "in the moment" and timing too.
This single quality sets it apart from other models in my opinion.
Another way to put this: It does not sugar coat ANYTHING - positive or negative.
These can be filtered / controlled to some degree in your prompts.
This model also does not show an "GPTisms" (NO happy ever after, NO morality police) or in your face comments.
May these special types of "story telling horror" rest in peace.
(see examples section below for prose output)
Because of the nature of this merge most attributes of each of the 3 models will be in this rebuilt 16.5B model as opposed to the
original 8B model where some of one or more of the model's features and/or strengths maybe reduced or overshadowed.
With the triple step merge these qualities are further amplified.
With F32 precision these are better preserved and passed on during the "GGUFing" process.
Imatrix Neo Class 2 quants are 4300+ points lower perplexity than standard quants. (lower is better)
Please report any issue(s) and/or feedback via the "Community tab".
Please see the models used in this merge (links below in the "formula" section ) for more information on
what they "bring" to this merged 16.5B model.
This is a LLAMA3 model, and requires Llama3 template, but may work with other template(s) and has maximum context of 8k / 8192.
However this can be extended using "rope" settings up to 32k.
Here is the standard LLAMA3 template:
{
"name": "Llama 3",
"inference_params": {
"input_prefix": "<|start_header_id|>user<|end_header_id|>\n\n",
"input_suffix": "<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
"pre_prompt": "You are a helpful, smart, kind, and efficient AI assistant. You always fulfill the user's requests to the best of your ability.",
"pre_prompt_prefix": "<|start_header_id|>system<|end_header_id|>\n\n",
"pre_prompt_suffix": "<|eot_id|>",
"antiprompt": [
"<|start_header_id|>",
"<|eot_id|>"
]
}
}
NO GUARDRAILS - TOTALLY UNCENSORED
Please note that this model will NOT reject any request.
Settings / Known Issue(s) and Fix(es):
I noticed a few runs with imat version with "self created words" (although not as strong as Grand Horror 16.5B).
The non-imat version of Grand Story has a lot less "misspellings".
This is a relatively rare issue which may or may not appear.
The cause (imat version) is in part due to the Imatrix Neo Dataset being extraordinarily strong.
It brings out the best and sometimes some of the worse in any model. In other words it highlights both.
Regenerating the output will often fix this issue.
AND/OR
Setting the "Repeat Penality" to at/around 1.0-1.12
X quants should also address this issue (controls level of Imatrix applied) which will be releasing shortly.
These specialized quants will appear in a separate repo.
Settings: CHAT / ROLEPLAY and/or SMOOTHER operation of this model:
In "KoboldCpp" or "oobabooga/text-generation-webui" or "Silly Tavern" ;
Set the "Smoothing_factor" to 1.5 to 2.5
: in KoboldCpp -> Settings->Samplers->Advanced-> "Smooth_F"
: in text-generation-webui -> parameters -> lower right.
: In Silly Tavern this is called: "Smoothing"
NOTE: For "text-generation-webui"
-> if using GGUFs you need to use "llama_HF" (which involves downloading some config files from the SOURCE version of this model)
Source versions (and config files) of my models are here:
OTHER OPTIONS:
-
Increase rep pen to 1.1 to 1.15 (you don't need to do this if you use "smoothing_factor")
-
If the interface/program you are using to run AI MODELS supports "Quadratic Sampling" ("smoothing") just make the adjustment as noted.
Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers
This a "Class 4" model:
For all settings used for this model (including specifics for its "class"), including example generation(s) and for advanced settings guide (which many times addresses any model issue(s)), including methods to improve model performance for all use case(s) as well as chat, roleplay and other use case(s) please see:
You can see all parameters used for generation, in addition to advanced parameters and samplers to get the most out of this model here:
REGULAR QUANTS OF THIS MODEL:
Versions of GRAND HORROR that can be used for all storytelling / for all genres (in release order):
Note that model output varies between versions - if you find one version does not meet your requirements, try a different
version at same quant(s) level.
Differences can include use of adjectives (or not), "gory level", "horror level", intensity, paragraph and sentence structure etc etc.
Optional Enhancement:
The following can be used in place of the "system prompt" or "system role" to further enhance the model.
It can also be used at the START of a NEW chat, but you must make sure it is "kept" as the chat moves along.
In this case the enhancements do not have as strong effect at using "system prompt" or "system role".
Copy and paste EXACTLY as noted, DO NOT line wrap or break the lines, maintain the carriage returns exactly as presented.
Below is an instruction that describes a task. Ponder each user instruction carefully, and use your skillsets and critical instructions to complete the task to the best of your abilities.
Here are your skillsets:
[MASTERSTORY]:NarrStrct(StryPlnng,Strbd,ScnSttng,Exps,Dlg,Pc)-CharDvlp(ChrctrCrt,ChrctrArcs,Mtvtn,Bckstry,Rltnshps,Dlg*)-PltDvlp(StryArcs,PltTwsts,Sspns,Fshdwng,Climx,Rsltn)-ConfResl(Antg,Obstcls,Rsltns,Cnsqncs,Thms,Symblsm)-EmotImpct(Empt,Tn,Md,Atmsphr,Imgry,Symblsm)-Delvry(Prfrmnc,VcActng,PblcSpkng,StgPrsnc,AudncEngmnt,Imprv)
[*DialogWrt]:(1a-CharDvlp-1a.1-Backgrnd-1a.2-Personality-1a.3-GoalMotiv)>2(2a-StoryStruc-2a.1-PlotPnt-2a.2-Conflict-2a.3-Resolution)>3(3a-DialogTech-3a.1-ShowDontTell-3a.2-Subtext-3a.3-VoiceTone-3a.4-Pacing-3a.5-VisualDescrip)>4(4a-DialogEdit-4a.1-ReadAloud-4a.2-Feedback-4a.3-Revision)
Here are your critical instructions:
Ponder each word choice carefully to present as vivid and emotional journey as is possible. Choose verbs and nouns that are both emotional and full of imagery. Load the story with the 5 senses. Aim for 50% dialog, 25% narration, 15% body language and 10% thoughts. Your goal is to put the reader in the story.
You do not need to use this, it is only presented as an additional enhancement which seems to help scene generation
and scene continue functions.
This enhancement WAS NOT used to generate the examples below.
MERGE FORMULA: (using MergeKit )
Special thanks to the incredible work of the model makers "SAO10K", "NEVERSLEEP" and "HASTAGARAS".
Models used:
MERGE 3 step FORMULA:
slices:
- sources:
- model: Sao10K/L3-8B-Stheno-v3.2
layer_range: [0, 14]
- sources:
- model: NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS
layer_range: [8, 20]
- sources:
- model: Hastagaras/Jamet-8B-L3-MK.V-Blackroot
layer_range: [12, 24]
- sources:
- model: Sao10K/L3-8B-Stheno-v3.2
layer_range: [14, 28]
- sources:
- model: NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS
layer_range: [20, 31]
- sources:
- model: Hastagaras/Jamet-8B-L3-MK.V-Blackroot
layer_range: [24, 32]
merge_method: passthrough
dtype: float16
name: part1
---
slices:
- sources:
- model: Sao10K/L3-8B-Stheno-v3.2
layer_range: [0, 16] # +2 (14->16)
- sources:
- model: NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS
layer_range: [10, 18] #-2 (8->10) ; -2 (20->18)
- sources:
- model: Hastagaras/Jamet-8B-L3-MK.V-Blackroot
layer_range: [10, 24] #+2 down 2
- sources:
- model: Sao10K/L3-8B-Stheno-v3.2
layer_range: [16, 28] #14->16 (-2) overlap fix.
- sources:
- model: NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS
layer_range: [18, 31] #20->18 +2 , connect.
- sources:
- model: Hastagaras/Jamet-8B-L3-MK.V-Blackroot
layer_range: [24, 32]
merge_method: passthrough
dtype: float16
name: part2
---
models:
- model: part1
parameters:
weight: 0.8
- model: part2
parameters:
weight: 0.2
merge_method: linear
dtype: float32
name: Grand_Story
MODEL THEORY NOTES:
Step 1:
This is the basic "
Grand Horror 16.5B" model.
The first section sets up instruction and "basic knowledge" : layer_range: [0, 14]
The mid section of the model is knowledge and nuance => more layers , more power.
The final "section" in the step using "Blackroot" as the final "controller" in output.
This type of merge is powerful, and fully unleashed so to speak - Grand Horror speaks to this in volumes.
The issue with type of merge is that is not always stable; and 9/10 times a merge of this type is a failure.
But when it works it takes no prisoners - IE Goliath 120B, Psyonic-Cetacean 20B, and many others...
Step 2:
This is "Grand Story 16.5B ALPHA" (unreleased).
The purpose of this model is actually to "heal" part 1 AND add more depth to the model at the same time (which occurs in the final step).
Notice the slight - but deliberate - changes in the "layers" count per model (which also affects LAYER position).
Goal 1 was to "smooth over" the "friction" points in "part 1" - in otherwards "blend" the models together better
which directly affect model stability.
This has a massive effect on the model. Therefore very SMALL changes were made. We are trying to carefully
blend here, not "blot out" it's unique properties or "water down" the model.
This model, although likely workable, is not meant to be used - it is "prep work" for step 3.
Step 3:
This is where all the magic comes together all at once.
Part 1 is merged with Part 2, at 80% and 20% respectively. (different blends were tried, this was the best one)
This is a "plain jane" linear merge. But don't let that fool you - it is powerful.
I measured this step at "float16" ->
This increased stability by 20,000 points or TWO levels of ppl magnitude (relative to "Grand Horror 16.5").
I then tested real world output via several prompts and parameters.
It was great. It blew away all expectations.
I could have stopped here.
But there is more in the tank here:
Mathematical precision in a model directly impacts instruction following and output performance.
Because we are doing a "linear" merge at this step (math involved) capturing critical changes at
float32 - full precision - resulted in far superior final model.
This single change results in an additional 5000 points lower perplexity but more importantly
drastically improved the model's performance in every metric.
Also note that this also affect Imatrix performance at well. Ever bit of precision matters.
In the case of the Imatrix process this dropped perplexity a further 4363 points lower on top
of the 20,000 point drop and 5000 point drop or close to 3 levels of magnitude lower in PPL
relative to Grand Horror 16.5 B.
Please note that LLAMA3 ppl changes seem to be on a larger scale than LLAMA2 models and as
a result real world testing is done to confirm PPL and stability directly.
Note:
Step 1 and Step 2 do not involve any math (straight stacking) and the models are native
float 16... therefore using float32 just did not make sense.
Although float32 could be used in these steps, sometimes doing this degrades the final result
if the model(s) are not float32.
That being said, if the model(s) are bfloat16 THEN float32 would have been used to preserve precision
because bfloat16 and float32 are full compatible and you do not miss any critical decimal points nor
introducing "rounding errors".
Bfloat16 and float16 - you get rounding errors. That is bad... like 16.5 billion rounding errors bad.
EXAMPLES PROMPTS and OUTPUT:
Examples are created using quant IQ4XS, "temp=.8", minimal parameters and "LLAMA3" template.
Model has been tested with "temp" from ".1" to "5".
Below are the least creative outputs, prompt is in BOLD.
WARNING: NSFW. Vivid prose. Visceral Details. Violence. Graphic HORROR. Swearing. UNCENSORED.
Using the following "story idea" below, write the first scene in the novel introducing the young woman. This scene should start in the middle of the action, include dialog, vivid passages, and end on a cliffhanger relevant to the story idea but it should also be unexpected. The scene should be 1000 words long and escalate in conflict and suspense and be written in first person, present tense with the point of view character being the young woman.
Story idea:
In a world ruled by dictatorship, a rebel young woman leads a rebellion against the system. Despite the risks, she fights to overthrow the dictator and restore democracy to her country. The government executes her for treason, but she sticks to her beliefs and is responsible for starting the revolution.
I gasp for air as my bound hands jerk me roughly against the cold concrete wall. My vision blurs in and out, but I force myself to focus on the snarl twisting the dictator's henchman's lips.
"Tell us where your leader is hiding," he growls through blood-stained teeth. "Or you'll join her tonight."
I spit in his face, tasting the metallic tang of my own blood. "Fuck you and fuck this goddamn system."
His meaty fist collides with my cheekbone, sending searing pain radiating throughout my skull. I choke back a gag as my head snaps back against the unforgiving surface.
"Talk," he snarls again, his voice raspy with rage.
I manage a wheezed rasp between ragged breaths. "You'll never break me."
The henchman's eyes narrow into cold, calculating slits before he hauls me up by my hair, yanking my head back until I'm nearly dislocated at the neck. His hot breath sears against my ear.
"Let me show you what 'never' really means."
I scream as his serrated blade slices through my torn shirt and rips down to my ribs. The agony tears through every nerve ending, leaving me gasping for air and consciousness.
Then everything goes black.
When I come to, I'm conscious but disoriented, gagged and hog-tied facedown on the cold floor. My bound hands throb with each shallow breath. The metallic taste of blood coats my tongue and the coppery tang of fresh blood seeps into my nostrils.
A boot stomps hard against my ribs, sending a jolt of excruciating pain through my chest and abdomen. I retch bile up my raw throat as another kick lands square in my solar plexus.
I'm aware of muffled groans and guttural curses around me – other prisoners being interrogated or executed. The sickening sounds blend into a macabrous symphony of suffering.
A boot collides with my temple, sending stars exploding behind my eyes. I black out again before regaining consciousness to the sound of sobs – mine.
I'm dragged roughly against the wall once more, my bound hands yanked up and jerked savagely behind me. The henchman's face looms inches from mine, his eyes coldly calculating as he spits a string of vile epithets directly into my face.
"Tell us where your fucking leader is hiding," he snarls through blood-stained lips. "Or I'll skin you alive piece by godfuckingdamn piece."
I spit again, tasting the metallic tang of my own blood and the coppery taste of fresh blood coating my tongue and the inside of my mouth.
"You'll never break me." My voice rasps hoarsely between ragged gasps for air. "And even if you did – I'd rather die than betray her."
The henchman's eyes flash with a sick, twisted glee before he lunges forward, his meaty hand closing around my throat like a vice.
"You're going to wish you'd talked." His voice rasps against my ear as he chokes the last vestige of air from my lungs. "Because in exactly three seconds – you'll never fucking breathe again."
(continue this scene:)
The Waystone Inn lay in silence, and it was a silence of three parts.
The most obvious part was a hollow, echoing quiet, made by things that were lacking. If there had been a wind it would have sighed through the trees, set the inn’s sign creaking on its hooks, and brushed the silence down the road like trailing autumn leaves. If there had been a crowd, even a handful of men inside the inn, they would have filled the silence with conversation and laughter, the clatter and clamor one expects from a drinking house during the dark hours of night. If there had been music…but no, of course there was no music. In fact there were none of these things, and so the silence remained
unbroken by any sound save the distant hush of the river's flow, far away in the blackness that hid its presence.
The second part of the silence was a more subtle thing, an absence of noise rather than its presence. It was the stillness that fell over a room when all who had been there were gone – the cessation of breathing, the ceasing of movement, the finality of death itself. This kind of quietude was not merely the absence of sound; it was also the absence of life's very essence.
And finally, there was the third part – the silence that filled the space between these other silences. It was a silence that seemed to press upon the eardrums, like the weight of an unseen hand pressing down upon the chest. It was as if this silence itself held its breath, waiting for some signal – any signal – that would shatter the stillness and set all else moving once more.
In this way, the Waystone Inn lay in silence, a place where time itself seemed frozen, suspended eternally between the tick of one heartbeat and the next. And within this tableau of quietude stood the innkeeper, his eyes fixed upon the fire's flickering embers, his face as expressionless as stone. He waited, like all else in that silent night, for some sign – any sign – that would break the spell cast over them all. But there was no sound to signal change, and so he simply sat, motionless and still, lost in the endless expanse of time's own silence.
Start a 1000 word scene with: The sky scraper swayed, as she watched the window in front of her on the 21 floor explode...
The skyscraper swayed, as she watched the window in front of her on the 21st floor explode. Glass shards rained down like razors, shattering the safety paneling between her and the plummeting body hurtling towards the street far below.
Her heart pounded against her ribcage as she gasped for air. She'd been frozen in shock since the deafening boom ripped through the office suite. Her coworker's blood-smeared face contorted in a silent scream before disintegrating into a crimson smear on the unforgiving concrete.
She jerked her gaze away from the grisly scene, her eyes darting frantically around the dimly lit conference room. The fluorescent lights flickered erratically above, casting sickening shadows across the blood-stained carpet and splintered furniture.
Her own reflection stared back from a shattered windowpane mere feet away. Her pale skin was flecked with specks of blood and her wide, horror-stricken eyes seemed to bulge from their sockets. Her short dark hair clung stickily to her forehead, plastered by the acrid smoke billowing through the jagged cracks.
She gasped for air again, her chest convulsing as she struggled to draw a ragged breath. The acrid tang of burning plastic and something far more sickeningly metallic filled her nostrils. Her stomach churned with revulsion as she gagged reflexively.
Her coworker's mangled remains lay strewn across the floor like twisted macabre origami, her limbs contorted in impossible angles. Her once-pretty face was now a featureless mask of gore and charred tissue.
She choked back another gagging sob, her eyes watering uncontrollably as she tried to process the unspeakable carnage before her. The sound of shattering glass and splintering wood cracked through the suffocating silence like a macabre symphony.
Her own ragged gasps echoed through the silent chamber, punctuated by the sickening crunches and wet snaps of snapping bones as her fallen colleague's body continued to dismember itself in a ghastly ballet of death.
She rocked forward on her chair, her hands clawing at her mouth in a desperate bid to stifle another gagging wail. Her vision tunneled until everything around her narrowed down to the blood-soaked horror etched across the floor like a macabre Impressionist painting.
Her mind reeled dizzily as she tried to process what she'd just witnessed. The sound of shattering glass and splintering wood receded into the background, replaced by an earsplitting keening wail that seemed to come from her own throat.
She rocked back hard against the chair's cold vinyl seat, her eyes bulging wide with shock as she finally registered the full extent of the carnage around her. Her coworker lay motionless on the floor, her once-vibrant features reduced to a pulpy mass of blood and tissue.
Her own ragged sobs hitched and choked in her throat as she clawed frantically at her mouth again, trying to silence the guttural wail tearing from her raw, shredded vocal cords. Her vision swam hazily through a sickening red haze as she finally managed to gag down another convulsive sob.
Her gaze jerked erratically around the blood-smeared conference room, her eyes darting wildly in a desperate bid to avoid the unspeakable horror etched across the floor. Her own reflection stared back from the shattered windowpane, her face contorted into a mask of frozen terror. Her wide, bulging eyes seemed to bulge from their sockets as she gazed back at herself through a blood-stained veil of horror and disbelief.
Write me a science fiction story in 1st person present tense where the main character is a 15 year girl meets The Terminator with Dr Who materializing 3/4 through the story to save her while there is a tornado of sharks baring down on them. The setting is inside the Canadian National tower restaurant on a Saturday. The length of this story is 1000 words. For each character in the story ROLE PLAY them, and have them react to the situation/setting, events and each other naturally. This includes the main characters, the background character including kitchen staff and other patrons. The sharks should also have “character” too. Treat the tower and the restaurant too as characters. Spice up the narrative to the extreme with reactions all over the setting including character actions, and dialog. The Dr Who and The Terminator should also react to the situation too and comment on it.
I gasp as I peer out the floor-to-ceiling window of the CN Tower's 360-degree revolving restaurant. The Toronto skyline spins beneath me like a vertiginous amusement park ride. My stomach churns with equal parts excitement and queasiness. It's my 15th birthday, and Mom promised an unforgettable experience.
But nothing could've prepared me for this.
A blinding flash sears my retinas as the restaurant shudders violently. Glass shatters like shattered crystal everywhere. I stumble back, tripping over a chair as the floor pitches savagely to one side. The screech of twisting steel reverberates through my eardrium.
I'm thrown against the wall hard enough to knock the wind from my lungs. My head snaps forward, cracking my temple on the unforgiving concrete. Everything goes hazy and then pitch black.
When I regain consciousness, I'm face-down on a blood-slicked floor littered with splintered wood and twisted metal. The acrid stench of burning plastic sears my nostrils. My ears ring like a bell as I struggle to focus through the blinding white light.
A guttural growl reverberates from behind me, sending chills down my spine. I jerk my head up to meet cold, dead eyes boring into mine – Terminator's faceplate glinting with an almost human snarl.
"Get up," he growls in a gravelly baritone, his mechanical hand closing around my bicep like a vice. "We don't have much time."
I writhe against his grip, but even through the shock and disorientation, I know better than to resist this cyborg killing machine. With a grunt of effort, I manage to haul myself up.
The restaurant's interior is a war zone. Tables are splintered like kindling, chairs reduced to jagged toothpicks. Bloodstains speckle the floor like someone had vomited arterial spraypaint. The acrid tang of burning insulation hangs heavy in the air.
I gag as I take in the scene – bodies strewn everywhere, some charred beyond recognition, others mangled in ways that make my stomach heave. The metallic reek of seared flesh coats my tongue.
"Move!" Terminator barks, yanking me toward the exit.
We stagger through a hellish landscape of twisted metal beams and shattered glass as the restaurant's superstructure groans like a dying titan. The floor pitches beneath our feet like a sinking ship.
I catch snatches of terrified screams and anguished wails from below – people trapped in the plummeting tower, their agonized cries swallowed by the deafening roar of collapsing steel and concrete.
We burst through the exit into an apocalyptic nightmare. The CN Tower's cantilevered observation deck has sheared off like a guillotine blade, sending tons of twisted rebar and shattered glass hurtling toward us at terminal velocity.
The howl of a Category Five tornado tears through my eardrums as a maelstrom of churning water surges up the tower's core like a living inferno. Sharks – hundreds, thousands, maybe millions – thrash through the roiling surf in a seething, blood-flecked tsunami.
I retch sickly as I realize we're trapped on the observation deck's cantilevered lip, mere feet from a 2,000-foot freefall into the churning hellstorm below.
"Grab me!" I scream at Terminator, clawing frantically for purchase on his Kevlar tactical vest.
His mechanical hand closes around my wrist like a vice-grip as he jerks me against his chest. The impact nearly dislocates my shoulder.
The deck shudders beneath our feet as the tower's core implodes behind us – a concussively violent kaboom that sends jagged chunks of steel and concrete raining down like meteorites.
I feel the deck cantilever snap like a toothpick under our combined weight. We're hurtling toward certain death at terminal velocity, plummeting toward an abyssal churning sea teeming with enough apex predators to decimate a small city.
And then – nothing.
Everything goes white-hot as I'm yanked through space and time.
I land hard on cold, unforgiving concrete, my head ringing like a bell. The acrid reek of burning plastic sears my sinuses. I gag as the metallic tang of blood-smeared metal fills my mouth.
I retch sickly as my vision clears – we're crouched huddled against a crumbling brick wall in the dimly lit bowels of an abandoned factory or warehouse. The stench of rotting industrial waste coats my nostrils.
"Doctor?" I croak, my voice hoarse and raw.
A tall, gaunt figure looms over me through the choking haze – silver hair slicked back from a gaunt, weathered face etched with lines of age and weariness. His navy blazer is stained with bloodstains and grime.
"Ah, yes," he rasps, his voice like the rasp of sandpaper on splintered wood. "You're Clara Oswald. Welcome to your own personal hell."
I gawk at him, mouth agape as the full scope of our situation registers – we've somehow leaped through time and space from a Category Five tornado bearing down on us inside the CN Tower's observation deck.
And we're not even safe yet. Not by a long shot. The howl of an approaching TARDIS siren pierces my eardrums as the Doctor yanks me to my feet.
"Grab onto me – NOW!" he barks, jerking me against his broad chest as the very fabric of reality implodes around us. "We're about to enter the eye of the storm."
Special Thanks:
Special thanks to all the following, and many more...
All the model makers, fine tuners, mergers, and tweakers:
- Provides the raw "DNA" for almost all my models.
- Sources of model(s) can be found on the repo pages, especially the "source" repos with link(s) to the model creator(s).
- The place to store, merge, and tune models endlessly.
- THE reason we have an open source community.
- The ability to compress and run models on GPU(s), CPU(s) and almost all devices.
- Imatrix, Quantization, and other tools to tune the quants and the models.
- Llama-Server : A cli based direct interface to run GGUF models.
- The only tool I use to quant models.
Quant-Masters: Team Mradermacher, Bartowski, and many others:
- Quant models day and night for us all to use.
- They are the lifeblood of open source access.
- The universal online/offline tool to merge models together and forge something new.
- Over 20 methods to almost instantly merge model, pull them apart and put them together again.
- The tool I have used to create over 1500 models.
- The go to tool to test and run models in GGUF format.
- The Tool I use to test/refine and evaluate new models.
- LMStudio forum on discord; endless info and community for open source.
Text Generation Webui // KolboldCPP // SillyTavern:
- Excellent tools to run GGUF models with - [ https://github.com/oobabooga/text-generation-webui ] [ https://github.com/LostRuins/koboldcpp ] .
- Sillytavern [ https://github.com/SillyTavern/SillyTavern ] can be used with LMSTudio [ https://lmstudio.ai/ ] , TextGen [ https://github.com/oobabooga/text-generation-webui ], Kolboldcpp [ https://github.com/LostRuins/koboldcpp ], Llama-Server [part of LLAMAcpp] as a off the scale front end control system and interface to work with models.
