Text Generation
Safetensors
GGUF
English
nae1
neuralai
causal-lm
llama-cpp
from-scratch
pretraining
sft
tool-use
local-ai
tiny-model
conversational
Instructions to use Subject-Emu-5259/NeuralAI-Nae1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use Subject-Emu-5259/NeuralAI-Nae1 with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf Subject-Emu-5259/NeuralAI-Nae1:Q4_K_M # Run inference directly in the terminal: llama cli -hf Subject-Emu-5259/NeuralAI-Nae1:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf Subject-Emu-5259/NeuralAI-Nae1:Q4_K_M # Run inference directly in the terminal: llama cli -hf Subject-Emu-5259/NeuralAI-Nae1:Q4_K_M
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-app # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama serve -hf Subject-Emu-5259/NeuralAI-Nae1:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama cli -hf Subject-Emu-5259/NeuralAI-Nae1:Q4_K_M
Use Docker
docker model run hf.co/Subject-Emu-5259/NeuralAI-Nae1:Q4_K_M
- Llama new
- LM Studio
- Jan
- vLLM
How to use Subject-Emu-5259/NeuralAI-Nae1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Subject-Emu-5259/NeuralAI-Nae1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Subject-Emu-5259/NeuralAI-Nae1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/Subject-Emu-5259/NeuralAI-Nae1:Q4_K_M
- Ollama
How to use Subject-Emu-5259/NeuralAI-Nae1 with Ollama:
ollama run hf.co/Subject-Emu-5259/NeuralAI-Nae1:Q4_K_M
- Unsloth Desktop
- Docker Model Runner
How to use Subject-Emu-5259/NeuralAI-Nae1 with Docker Model Runner:
docker model run hf.co/Subject-Emu-5259/NeuralAI-Nae1:Q4_K_M
- Lemonade
How to use Subject-Emu-5259/NeuralAI-Nae1 with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull Subject-Emu-5259/NeuralAI-Nae1:Q4_K_M
Run and chat with the model
lemonade run user.NeuralAI-Nae1-Q4_K_M
List all available models
lemonade list
- Atomic Chat
step-32000 refresh: PPL 54.16 (corrected-label cycle), weights/eval/charts/badges + fresh-corpus cross-check
Browse files- README.md +17 -13
- eval_fresh_step28000_q4_n512.json +18 -0
- eval_fresh_step32000_q4_n512.json +18 -0
- eval_step32000_q4_n512.json +18 -0
- nae1-llama-Q4_K_M.gguf +1 -1
- nae1-llama-f32.gguf +1 -1
- neuralai-nae1-banner.png +2 -2
- neuralai-nae1-banner.svg +1 -1
- neuralai-nae1-comparison.png +0 -0
- neuralai-nae1-comparison.svg +277 -254
- neuralai-nae1-training.png +2 -2
- neuralai-nae1-training.svg +10 -10
README.md
CHANGED
|
@@ -18,7 +18,7 @@ pipeline_tag: text-generation
|
|
| 18 |
|
| 19 |
<!--
|
| 20 |
NeuralAI — Nae1
|
| 21 |
-
Model card maintained by De'Andrew Preston Harris (@Subject-Emu-5259) Last updated: 2026-10-08 (step-
|
| 22 |
|
| 23 |
<img src="neuralai-nae1-banner.png" alt="NeuralAI Nae1 banner" width="100%" />
|
| 24 |
|
|
@@ -32,8 +32,8 @@ pipeline_tag: text-generation
|
|
| 32 |
<p align="center">
|
| 33 |
<a href="https://huggingface.co/Subject-Emu-5259/NeuralAI-Nae1"><img src="https://img.shields.io/badge/🤗%20Hub-NeuralAI%20Nae1-blue" alt="Nae1"></a>
|
| 34 |
<img src="https://img.shields.io/badge/params-100.1M-22d3ee" alt="100.1M params">
|
| 35 |
-
<img src="https://img.shields.io/badge/held--out%20PPL-
|
| 36 |
-
<img src="https://img.shields.io/badge/steps-
|
| 37 |
<img src="https://img.shields.io/badge/license-Apache%202.0-orange" alt="Apache 2.0">
|
| 38 |
</p>
|
| 39 |
|
|
@@ -49,9 +49,9 @@ pipeline_tag: text-generation
|
|
| 49 |
| **Vocabulary** | 439 tokens (ByteLevel BPE, sliced from a 32,000-target config) |
|
| 50 |
| **Context** | 2,048 trained · evaluated on `n_ctx=512` |
|
| 51 |
| **Training data** | 768M tokens — FineWeb CC-MAIN-2013-20, 400k documents, 8 shards |
|
| 52 |
-
| **Training schedule** |
|
| 53 |
-
| **Final training loss** |
|
| 54 |
-
| **Held-out perplexity** | **
|
| 55 |
| **Formats in this repo** | `f32` GGUF (305 MB) · `Q4_K_M` GGUF (45 MB) · config + tokenizer |
|
| 56 |
| **License** | Apache 2.0 |
|
| 57 |
|
|
@@ -60,10 +60,10 @@ pipeline_tag: text-generation
|
|
| 60 |
## 📈 Training
|
| 61 |
|
| 62 |
<p align="center">
|
| 63 |
-
<img src="neuralai-nae1-training.png" alt="Nae1 training loss step
|
| 64 |
</p>
|
| 65 |
|
| 66 |
-
Nae1 is pretrained from random initialization in
|
| 67 |
|
| 68 |
- **Data pipeline**: FineWeb parquet → chunked into 513-token windows served as zero-copy views (peak-RAM-safe at 768M tokens)
|
| 69 |
- **Checkpoints**: every 250 steps; this release ships the final step-28000 checkpoint
|
|
@@ -88,9 +88,10 @@ Teacher-forced perplexity on a held-out FineWeb slice (992 documents / 1,027,766
|
|
| 88 |
| step-16000 | 151.45 |
|
| 89 |
| step-20000 | 148.84 |
|
| 90 |
| step-24000 | 147.97 |
|
| 91 |
-
|
|
|
|
|
| 92 |
|
| 93 |
-
Raw report: [`
|
| 94 |
|
| 95 |
> ⚠️ Perplexity is the trust signal here, not raw capability. At 100M params over 768M tokens, expect exploratory, often garbled text — this is a research-lineage model, not a chat assistant.
|
| 96 |
|
|
@@ -131,7 +132,10 @@ print(out["choices"][0]["text"])
|
|
| 131 |
| `nae1-llama-f32.gguf` | Full-precision GGUF, 305 MB — for requantization / research |
|
| 132 |
| `config.json` | Architecture config as trained |
|
| 133 |
| `tokenizer/` | Native 439-token vocab + merges + manifest |
|
| 134 |
-
| `
|
|
|
|
|
|
|
|
|
|
| 135 |
| `eval_step24000_q4_n512.json` | Prior release (step-24000), kept for lineage |
|
| 136 |
| `eval_step20000_q4_n512.json` | Prior release (step-20000), kept for lineage |
|
| 137 |
| `eval_step16000_q4_n512.json` | Prior release (step-16000), kept for lineage |
|
|
@@ -149,7 +153,7 @@ print(out["choices"][0]["text"])
|
|
| 149 |
|
| 150 |
- **Scale**: 100M parameters pretrained on 768M tokens is a research checkpoint — long-form reasoning, coding, and factual recall are limited.
|
| 151 |
- **Vocabulary**: a 439-token BPE trained on this corpus; out-of-domain text will tokenize inefficiently.
|
| 152 |
-
- **No chat template**: base completion only — it was never instruction-tuned (SFT
|
| 153 |
- **No internet access**: pair with a tool layer if you need live data.
|
| 154 |
|
| 155 |
---
|
|
@@ -175,7 +179,7 @@ NeuralAI was built from resilience, fatherhood, and the belief that personal com
|
|
| 175 |
title = {NeuralAI — Nae1},
|
| 176 |
year = {2026},
|
| 177 |
url = {https://huggingface.co/Subject-Emu-5259/NeuralAI-Nae1},
|
| 178 |
-
version = {step-
|
| 179 |
description = {A 100M-parameter language model pretrained from scratch on 768M FineWeb tokens}
|
| 180 |
}
|
| 181 |
```
|
|
|
|
| 18 |
|
| 19 |
<!--
|
| 20 |
NeuralAI — Nae1
|
| 21 |
+
Model card maintained by De'Andrew Preston Harris (@Subject-Emu-5259) Last updated: 2026-10-08 (step-32000 refresh)-->
|
| 22 |
|
| 23 |
<img src="neuralai-nae1-banner.png" alt="NeuralAI Nae1 banner" width="100%" />
|
| 24 |
|
|
|
|
| 32 |
<p align="center">
|
| 33 |
<a href="https://huggingface.co/Subject-Emu-5259/NeuralAI-Nae1"><img src="https://img.shields.io/badge/🤗%20Hub-NeuralAI%20Nae1-blue" alt="Nae1"></a>
|
| 34 |
<img src="https://img.shields.io/badge/params-100.1M-22d3ee" alt="100.1M params">
|
| 35 |
+
<img src="https://img.shields.io/badge/held--out%20PPL-54.16-34d399" alt="PPL 54.16">
|
| 36 |
+
<img src="https://img.shields.io/badge/steps-32,000-a78bfa" alt="32000 steps">
|
| 37 |
<img src="https://img.shields.io/badge/license-Apache%202.0-orange" alt="Apache 2.0">
|
| 38 |
</p>
|
| 39 |
|
|
|
|
| 49 |
| **Vocabulary** | 439 tokens (ByteLevel BPE, sliced from a 32,000-target config) |
|
| 50 |
| **Context** | 2,048 trained · evaluated on `n_ctx=512` |
|
| 51 |
| **Training data** | 768M tokens — FineWeb CC-MAIN-2013-20, 400k documents, 8 shards |
|
| 52 |
+
| **Training schedule** | 32,000 steps — eight cosine cycles (0→4k→…→28k→32k) · Kaggle T4 |
|
| 53 |
+
| **Final training loss** | 17.93 @ step 32,000 (sum over 8 grad-accum micros; cycle 8 is the first trained on corrected next-token labels — resume started at 28.87 under the fixed objective, so pre-fix values are not comparable) |
|
| 54 |
+
| **Held-out perplexity** | **54.16** (Q4_K_M, 992 docs / 1,027,766 tokens, teacher-forced; cross-checked **57.09** on a fresh unseen FineWeb slice vs step-28000's 143.64 — the improvement is real generalization) |
|
| 55 |
| **Formats in this repo** | `f32` GGUF (305 MB) · `Q4_K_M` GGUF (45 MB) · config + tokenizer |
|
| 56 |
| **License** | Apache 2.0 |
|
| 57 |
|
|
|
|
| 60 |
## 📈 Training
|
| 61 |
|
| 62 |
<p align="center">
|
| 63 |
+
<img src="neuralai-nae1-training.png" alt="Nae1 training loss step 28000 to 32000" width="92%" />
|
| 64 |
</p>
|
| 65 |
|
| 66 |
+
Nae1 is pretrained from random initialization in eight cosine cycles: 0 → 4,000 · 4,000 → 8,000 · … · 24,000 → 28,000 · 28,000 → 32,000, each resumed from the previous checkpoint. The chart shows cycle eight — the **first cycle trained with corrected next-token labels** (an earlier double-shift bug had models predicting token t+2; the held-out gate was always standard next-token, so historical PPL numbers remain valid). Under the corrected objective the logged loss resumes at 28.87, falls to a best of **15.72** (step 31,500) and finishes at **17.93** as the learning rate anneals to zero. Held-out PPL improves 147.97 → 146.17 → **54.16** — a 63% drop from fixing the training objective; a fresh-corpus cross-check (unseen FineWeb CC-MAIN-2014-10: **57.09** vs step-28000's **143.64**) confirms the jump is genuine generalization, not eval-set leakage.
|
| 67 |
|
| 68 |
- **Data pipeline**: FineWeb parquet → chunked into 513-token windows served as zero-copy views (peak-RAM-safe at 768M tokens)
|
| 69 |
- **Checkpoints**: every 250 steps; this release ships the final step-28000 checkpoint
|
|
|
|
| 88 |
| step-16000 | 151.45 |
|
| 89 |
| step-20000 | 148.84 |
|
| 90 |
| step-24000 | 147.97 |
|
| 91 |
+
| step-28000 | 146.17 |
|
| 92 |
+
| **step-32000 — this release** | **54.16** |
|
| 93 |
|
| 94 |
+
Raw report: [`eval_step32000_q4_n512.json`](eval_step32000_q4_n512.json) — mean NLL 3.9919 nats/token; fresh-corpus cross-check: [`eval_fresh_step32000_q4_n512.json`](eval_fresh_step32000_q4_n512.json) / [`eval_fresh_step28000_q4_n512.json`](eval_fresh_step28000_q4_n512.json) (prior releases kept alongside: [`eval_step24000_q4_n512.json`](eval_step24000_q4_n512.json), [`eval_step20000_q4_n512.json`](eval_step20000_q4_n512.json), [`eval_step16000_q4_n512.json`](eval_step16000_q4_n512.json), [`eval_step8000_q4_n512.json`](eval_step8000_q4_n512.json)). Fixed probe ("The quick brown fox", 10 tokens): 174.73 — the 10-token probe is noisy; held-out decides.
|
| 95 |
|
| 96 |
> ⚠️ Perplexity is the trust signal here, not raw capability. At 100M params over 768M tokens, expect exploratory, often garbled text — this is a research-lineage model, not a chat assistant.
|
| 97 |
|
|
|
|
| 132 |
| `nae1-llama-f32.gguf` | Full-precision GGUF, 305 MB — for requantization / research |
|
| 133 |
| `config.json` | Architecture config as trained |
|
| 134 |
| `tokenizer/` | Native 439-token vocab + merges + manifest |
|
| 135 |
+
| `eval_step32000_q4_n512.json` | The exact evaluation report cited above |
|
| 136 |
+
| `eval_fresh_step32000_q4_n512.json` | Fresh-corpus cross-check (unseen CC-MAIN-2014-10), this release |
|
| 137 |
+
| `eval_fresh_step28000_q4_n512.json` | Fresh-corpus cross-check, prior release |
|
| 138 |
+
| `eval_step28000_q4_n512.json` | Prior release (step-28000), kept for lineage |
|
| 139 |
| `eval_step24000_q4_n512.json` | Prior release (step-24000), kept for lineage |
|
| 140 |
| `eval_step20000_q4_n512.json` | Prior release (step-20000), kept for lineage |
|
| 141 |
| `eval_step16000_q4_n512.json` | Prior release (step-16000), kept for lineage |
|
|
|
|
| 153 |
|
| 154 |
- **Scale**: 100M parameters pretrained on 768M tokens is a research checkpoint — long-form reasoning, coding, and factual recall are limited.
|
| 155 |
- **Vocabulary**: a 439-token BPE trained on this corpus; out-of-domain text will tokenize inefficiently.
|
| 156 |
+
- **No chat template**: base completion only — it was never instruction-tuned (SFT completed; checkpoint selection in progress).
|
| 157 |
- **No internet access**: pair with a tool layer if you need live data.
|
| 158 |
|
| 159 |
---
|
|
|
|
| 179 |
title = {NeuralAI — Nae1},
|
| 180 |
year = {2026},
|
| 181 |
url = {https://huggingface.co/Subject-Emu-5259/NeuralAI-Nae1},
|
| 182 |
+
version = {step-32000},
|
| 183 |
description = {A 100M-parameter language model pretrained from scratch on 768M FineWeb tokens}
|
| 184 |
}
|
| 185 |
```
|
eval_fresh_step28000_q4_n512.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model": "artifacts/nae1_gguf/nae1-llama-Q4_K_M-step28000.gguf",
|
| 3 |
+
"context_size": 512,
|
| 4 |
+
"scoring": "teacher forcing: one llm.eval(chunk) + row j scores token j+1 (first token excluded)",
|
| 5 |
+
"probe": {
|
| 6 |
+
"text": "The quick brown fox",
|
| 7 |
+
"tokens": 10,
|
| 8 |
+
"total_nll_nats": 51.63228580836413,
|
| 9 |
+
"perplexity": 174.72766832853299
|
| 10 |
+
},
|
| 11 |
+
"heldout": {
|
| 12 |
+
"documents": 300,
|
| 13 |
+
"tokens": 350788,
|
| 14 |
+
"total_nll_nats": 1742465.3284437277,
|
| 15 |
+
"mean_nll_nats_per_token": 4.967288870895605,
|
| 16 |
+
"perplexity": 143.63694072145765
|
| 17 |
+
}
|
| 18 |
+
}
|
eval_fresh_step32000_q4_n512.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model": "artifacts/nae1_gguf/nae1-llama-Q4_K_M-step32000.gguf",
|
| 3 |
+
"context_size": 512,
|
| 4 |
+
"scoring": "teacher forcing: one llm.eval(chunk) + row j scores token j+1 (first token excluded)",
|
| 5 |
+
"probe": {
|
| 6 |
+
"text": "The quick brown fox",
|
| 7 |
+
"tokens": 10,
|
| 8 |
+
"total_nll_nats": 34.301195067379204,
|
| 9 |
+
"perplexity": 30.880332937026726
|
| 10 |
+
},
|
| 11 |
+
"heldout": {
|
| 12 |
+
"documents": 300,
|
| 13 |
+
"tokens": 350788,
|
| 14 |
+
"total_nll_nats": 1418830.3483974792,
|
| 15 |
+
"mean_nll_nats_per_token": 4.044694654313942,
|
| 16 |
+
"perplexity": 57.093750042001716
|
| 17 |
+
}
|
| 18 |
+
}
|
eval_step32000_q4_n512.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model": "/Users/deandrewharris/Projects/NuralAI-Project-is-stored-on-zo-computer/artifacts/nae1_gguf/nae1-llama-Q4_K_M-step32000.gguf",
|
| 3 |
+
"context_size": 512,
|
| 4 |
+
"scoring": "teacher forcing: one llm.eval(chunk) + row j scores token j+1 (first token excluded)",
|
| 5 |
+
"probe": {
|
| 6 |
+
"text": "The quick brown fox",
|
| 7 |
+
"tokens": 10,
|
| 8 |
+
"total_nll_nats": 34.301195067379204,
|
| 9 |
+
"perplexity": 30.880332937026726
|
| 10 |
+
},
|
| 11 |
+
"heldout": {
|
| 12 |
+
"documents": 992,
|
| 13 |
+
"tokens": 1027766,
|
| 14 |
+
"total_nll_nats": 4102713.7013922515,
|
| 15 |
+
"mean_nll_nats_per_token": 3.9918752920336456,
|
| 16 |
+
"perplexity": 54.15635317413229
|
| 17 |
+
}
|
| 18 |
+
}
|
nae1-llama-Q4_K_M.gguf
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 45762464
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe1492ec04cc3744a7306e4303dffe25a7fb02f64cb24b7087a8c84ca8807704
|
| 3 |
size 45762464
|
nae1-llama-f32.gguf
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 304778816
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2eba24ceefa28a3da26d4d06c79fc478932ada989c3cd889ce1225510ee33465
|
| 3 |
size 304778816
|
neuralai-nae1-banner.png
CHANGED
|
Git LFS Details
|
|
Git LFS Details
|
neuralai-nae1-banner.svg
CHANGED
|
|
|
|
neuralai-nae1-comparison.png
CHANGED
|
|
neuralai-nae1-comparison.svg
CHANGED
|
|
|
|
neuralai-nae1-training.png
CHANGED
|
Git LFS Details
|
|
Git LFS Details
|
neuralai-nae1-training.svg
CHANGED
|
|
|
|