Baseline

TRAIN v0.01

Baseline. Predicts each outcome's historical base rate and ignores the token entirely. Every later version has to beat this to be worth anything.

Completedpreviously deployed

Definition

Model type

Baseline (averages)

Signals used

Market signals

Technical details

Random seed

1001

Config hash

bd3315fe1aabd63a

Created

05 Oct 02:55

Completed

05 Oct 02:55

Artifact

models/v0.01/model.json

sha256 a4fb2a4dc9da1d59…

Dataset & compute

Dataset

Solana Launch Dataset v1

60 snapshots · 12 launches · sha256 010016f628b3

Training runs

Compute

TRAIN training server (CPU)

GPU hours

—

no GPU used

CPU time

4.3 s

Provider cost

$0.00

nothing purchased

Benchmarks

held-out test set

Spotting crashes (−70% within 1h)

0.500

—

n=532 · 0.50–0.50

Spotting survivors (still traded after 1h)

0.500

—

n=532 · 0.50–0.50

Spotting 2× runs (within 1h)

0.500

—

n=532 · 0.50–0.50

Calling the 1h direction (down / flat / up)

33.3%

—

n=532 · 33–33

Spotting creator dumps (within 1h)

0.500

—

n=112 · 0.50–0.50

Spotting crashes (within 6h)

insufficient data

—

n=0

Spotting survivors (6h)

insufficient data

—

n=0

Calling the 6h direction

insufficient data

—

n=0

Spotting survivors (24h)

insufficient data

—

n=0

Spotting migrations (within 24h)

insufficient data

—

n=0

Validation metrics (during training)

collapse_1h

—

n=0

collapse_6h

—

n=0

reach_2x_1h

—

n=0

survival_1h

—

n=0

survival_6h

—

n=0

survival_24h

—

n=0

migration_24h

—

n=0

trajectory_1h

—

n=0

trajectory_6h

—

n=0

creator_exit_1h

—

n=0

Validation launches are separate from training launches and from the locked test set. Benchmarks above are the public numbers.

What this version relies on

held-out validation launches

This version was trained before TRAIN started measuring which signals each model relies on.

For each signal, how much the score drops when that signal is scrambled across launches. TRAIN learns these weights itself from the data; nobody hand-codes them.

Training runs

Attempt 1Succeeded

05 Oct 02:55 · 00:00:00 · 60 examples · $0.00

Timeline

05 Oct 06:05
TRAIN v0.01 not deployed; v0.01 stays live
Reason: .
05 Oct 02:55
TRAIN v0.01 deployed — now live
first completed model
05 Oct 02:55
Benchmark completed: TRAIN v0.01
No locked test set yet (not enough held-out launches with observed outcomes). Benchmarks will be computed when it is locked.
05 Oct 02:55
Benchmark started: TRAIN v0.01
05 Oct 02:55
Training completed: TRAIN v0.01
00:00:00 · trained on 60 examples · $0 compute cost
05 Oct 02:55
Training started: TRAIN v0.01
Solana Launch Dataset v1 · baseline (averages) · TRAIN training server
05 Oct 02:55
TRAIN v0.01 queued
Baseline. Predicts each outcome's historical base rate and ignores the token entirely. Every later version has to beat this to be worth anything.