Dataset Viewer
Auto-converted to Parquet Duplicate
text
stringlengths
700
12k
source
stringclasses
6 values
time_bucket
stringclasses
5 values
year
int64
1.95k
1.99k
approx_tokens
int64
175
3k
Aaymond G A Cote Senlor Editor Blaise W Liffick Edi ter Richard Shutord, N4ANG Editoria) Assistant Asalatant ae Oritton iw Pri Chee Newalatiers Laura A Hanson Ruth M Walsh Editor Adv/Prod Coordinator Thomas Harve Advertising Bill io Noreen Bardsley Jon Swanson Don Bardsiay 4 = June 1979 © BYTE Publica...
byte_magazine
1970s
1,979
2,988
“When I’m working on my programs tate at night. | can’t wait for cassette storage. My minifloppy gives me fast random access and data transfer. The little minidiskettes™ store plenty of data and file easily too. “| made the right decision when | bought a system with the minifloppy. When you lay out your own hard-ear...
byte_magazine
1970s
1,979
2,994
The brain is first and foremost a control system. All brains, even that of the tiniest insect, control behavior. Some brains can produce very complex behavior, but only the most sophisticated and highly developed brains exhibit the phenomenon of thought. Clearly then, thought is not the central purpose of the brain, bu...
byte_magazine
1970s
1,979
2,923
A vector in a higher dimensional space can usually be visualized as a projection onto a lower dimensional space. For example, typical mechancial drawings portray front, side, and top views of a three-dimensional form projected onto a two-dimensiona! sheet of paper. Each projection can either illustrate a cut through th...
byte_magazine
1970s
1,979
2,981
Figure 11: We will define the set of operators H = (h tion which maps the input vector § into the output vector P., p Ay. A,) as @ func- june 1979 © BYTE Publications tac = 17 INPUT SPACE OUTPUT SPACE Figure 12: The operator H maps every input vector S in input space into an output vector P in output space. H...
byte_magazine
1970s
1,979
2,980
Circte 66 on inquiry card. June 1979 © BYTE Publications Inc 21 cook up all your favorites. Now, we're cooking. Our boys in the lab have turned circuit chefs these past three months to create a smorgas- bord of deliciously assembied boards to support your APPLE II* TRS-80' or S-100 bus systems. Feast your eyes on o...
byte_magazine
1970s
1,979
2,992
OBSERVED OBSERVED VELOCITY F OBSERVED | POSITION POSITION OBSERVED FORCE, ETC 1 |verociry | ano FORCE iy ae MOTOR DRIVE SIGNALS TO ACTUATORS (MOTORS, VALVES, MUSCLES, ETC } eS Aes |e See caeR OUTPUT Ge OBSERVABLE GOAL-OIRECTED SENSORY-INTERACTIVE BEHAVIOR OF CREATURE IN ITS ENVIRONMENT VOLTAGES NERVE (PULSES, (a...
byte_magazine
1970s
1,979
2,973
TI’s new TM990/189 University Module is a stand-alone learning lab. Fully assembled and designed for maximum hands-on experience. To ease and simplify learning and teaching. Outstanding features include powerful 16-bit microprocessor with easy-to-learn, easy-to-use minicomputer instruction set; 45- key alphanumeric ke...
byte_magazine
1970s
1,979
2,980
a ee @ Instantaneous Formatting. Compacting (extraneous blank deletion) and right justifying are simple commands that tidy up a full page or window's worth of text in the blink File Switching. Moving from document to document to of an eye. Random access cursor movement, line and examine, copy, move and change text is...
byte_magazine
1970s
1,979
2,964
{0, 0) (0, 0) Figure 6: Maze pitfalls. On (10, 0} 10, 0} its first, nonoptimized agen a tea attempt, the mouse will cdo. Oth : traverse the entire length (30, 20) of the dead-end corridor. (20, 20) Pa > ca es After optimizing the path ie 4 Optimization pe I Nett eee Now that the mouse has a way of remem- hin e ea berin...
byte_magazine
1970s
1,979
2,970
path which has been tra- versed. Examples of short paths are the gray path from A to B, either path from B to C (since they | are the same}, and the red path from C to D. The | final path is the most Y direct and shortest path Yy through this section of the YW maze. A bw | x TOTAL LENGTH TOTAL LENGTH 710 UNIT...
byte_magazine
1970s
1,979
2,980
ing T/PM or other CP/M* like DOS, you can now run that pregram simply, without patches. Floppy Disk $149.95 *Trademark of Digital Research Circle 308 on inquiry card. PROGRAMS cee eee ee ee ee Se ee ee eR ee ee a 8 ee ew He ee eee ee See ee ee Ee ee ee 8 ee eee eee ee eee ewe Meee een SOFTWARE 1. MICROFILE: A...
byte_magazine
1970s
1,979
2,942
There fave been many articles on the subject of alpha brain wave and muscle monitors; some even include circuit diagrams for construction of the interfaces. The major thing these articles lack is a caution about matching components, and the critical importance of proper layout. The circuit of figure 1, if breadboarded ...
byte_magazine
1970s
1,979
2,986
ciTY STATE ZIP. DAYTIME PHONE TITLE OF PROGRAM CATEGORY C Business © Fun & Games O Education O Home/Personal Management SIGNATURE DATE Copynght 1979, Exidy, Inc. Photo 5: The electrode has a saturated spongy center which serves to reduce skin contact resistance. It is necessary to use this type of connect...
byte_magazine
1970s
1,979
2,930
Join the thousands of InterTube celebrations going on around the country at this very moment. Call us at the number below and start your own celebration (BYOB—we'll bring the } dt UPe DATA = SYSEMS. 2300 Broad River Road, Columbia, S. C, 29210 BYTE June 1979 59 The sales literature for the Apple II lists the spe...
byte_magazine
1970s
1,979
2,925
End of preview. Expand in Data Studio

Echo88-150M-Base

Echo88-150M-Base is a small English decoder-only causal language model trained from scratch on the Echo88 pretraining dataset.

The goal of Echo88 is to create a compact base model inspired by the language, computing culture, printed media, Usenet discussion, and older book knowledge available up to the late 1980s.

This is a base model, not an instruction-tuned chatbot. It is trained for next-token prediction and should be fine-tuned before being used as a helpful assistant.

Model Details

Model name: Echo88-150M-Base
Model type: Decoder-only causal language model
Training type: From scratch
Approx size: 150M parameters
Language: English
Context length: 2048 tokens
Tokenizer: Echo88 custom tokenizer
Intended use: Base pretraining / text generation / further fine-tuning

Training Data

Echo88-150M-Base was trained on the Echo88 Base Dataset, a cleaned English text corpus of approximately 1.17B tokens.

The dataset includes:

Books / Gutenberg-style public-domain text
UTZOO Usenet posts
BYTE Magazine
PC Magazine
TIME Magazine
Internet Archive Magazine Rack OCR text
Computer and technology magazine text
General historical magazine text

The dataset is designed to emphasize the period from the 1950s through the late 1980s, with a strong focus on early personal computing, Usenet, printed magazines, and older long-form writing.

Related dataset:

guus4324343/Echo88-Pretrain-1.17B

Intended Use

This model is intended for:

causal language modeling
retro / historical AI experiments
small language model research
continued pretraining
instruction tuning
1980s-style assistant experiments
computer-history model experiments

Recommended next step:

Echo88-150M-Base
→ supervised fine-tune on Echo88-Instruct-173K
→ Echo88-150M-Instruct

Not Instruction Tuned

This model is not yet trained to follow instructions reliably.

For chat or assistant behavior, use or create an instruction-tuned version using:

guus4324343/Echo88-Instruct-173K

Expected behavior of the base model:

continues text
completes paragraphs
imitates source style
may produce raw text rather than direct answers
may not follow commands consistently

Knowledge Boundary

Echo88 is designed around a historical data mixture ending around the late 1980s.

The model should not be expected to know modern topics such as:

Google
Wikipedia
iPhone
smartphones
modern social media
Windows 95 and later software
COVID-19
modern AI systems
2000s/2010s/2020s events

Because this is a base model, it may still hallucinate if prompted about modern events. A later instruction-tuned model should be trained to respond more carefully to post-1988 topics.

Example Usage

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_id = "guus4324343/Echo88-150M-Base"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.float16,
    device_map="auto"
)

prompt = "The personal computer revolution of the 1980s"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)

with torch.no_grad():
    output = model.generate(
        **inputs,
        max_new_tokens=120,
        temperature=0.8,
        top_p=0.95,
        do_sample=True
    )

print(tokenizer.decode(output[0], skip_special_tokens=True))

Limitations

Echo88-150M-Base is experimental and small.

Known limitations:

not instruction tuned
may hallucinate
may repeat text
may produce OCR-like artifacts
may reflect outdated historical language or views
may struggle with complex reasoning
may not reliably refuse post-1988 topics
may produce incomplete or strange continuations

The model is intended for research and experimentation, not high-stakes use.

Bias and Historical Content

The training data includes historical books, magazines, and Usenet text. As a result, the model may reproduce outdated language, assumptions, stereotypes, or viewpoints present in older source material.

Users should review outputs carefully.

Training Notes

This model was trained as the base stage of the Echo88 project.

Planned model family:

Echo88-150M-Base
Echo88-150M-Instruct
Echo88-150M-Chat

License

The model weights are released under the Apache 2.0 license.

The training dataset is mixed-source and is released separately under other. Users are responsible for checking dataset source rights and suitability for their own use case.

Downloads last month
22

Models trained or fine-tuned on exnivo/Echo88-150M-Base

Collection including exnivo/Echo88-150M-Base