Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/alphaleaks60-maker/docs2/llms.txt

Use this file to discover all available pages before exploring further.

The standard model consumes a 68-feature vector assembled at inference time from multiple data sources: wallet history, signal-level event data, creator intelligence, token state, social graph scores, and live market context. Every feature is computed by the same code path in production as during training, ensuring no distribution shift between training and serving.
Features must be assembled in exactly the order defined by the model’s feature_names metadata array. The FEATURE_ORDER constant in the codebase is the canonical source of truth. Assembling features in a different order will not produce a runtime error — the model will silently generate incorrect predictions.

Wallet point-in-time (4 features)

These features capture the wallet’s stats as they were at the moment the signal fired — not their current values. The point-in-time snapshot is what prevents lookahead bias: the model only sees information that was actually available when the trade decision would have been made.
FeatureDescription
alpha_score0–100 composite quality score at time of signal
wallet_graduation_rateShare of tokens bought that graduated, at signal time
wallet_avg_buy_rankTypical buy rank at signal time (lower = earlier entry)
wallet_trades_atTotal trades observed at signal time

Wallet behavioural (15 features)

These features describe the wallet’s trading behaviour across its full observed history. They are computed by FeatureComputer on a 30-minute rolling basis for all active wallets.
FeatureDescription
tokens_tradedTotal unique tokens bought
pct_top10_entriesFraction of buys that were in the first 10 wallets
avg_seconds_after_creationAverage delay between token creation and the wallet’s buy
pct_buys_under_60sFraction of buys within 60 seconds of token creation
pct_buys_under_300sFraction of buys within 300 seconds of token creation
avg_sol_per_buyMean buy size in SOL
median_sol_per_buyMedian buy size in SOL
sol_size_stddevStandard deviation of buy sizes (high = inconsistent sizing)
max_single_buyLargest single buy ever recorded
avg_hold_time_secondsAverage time between buy and sell
pct_quick_flipsFraction of positions closed within 5 minutes
pct_diamond_handsFraction of positions held over 24 hours
unique_creators_tradedNumber of distinct token creators the wallet has engaged with
pct_repeat_creator_buysFraction of buys on tokens from creators the wallet has bought before
active_days_30dDistinct active trading days in the last 30 days

Wallet returns (11 features)

Computed from PnlCalculator and PeakTracker output. These features describe how profitable the wallet has been historically — both in terms of realised exits and peak opportunity captured.
FeatureDescription
graduation_rateShare of bought tokens that graduated
win_rateShare of closed positions that exited in profit
sol_weighted_returnSOL-weighted average return across all positions
avg_realized_multipleAverage exit multiple
avg_peak_multipleAverage of peak prices seen during the hold period
capture_efficiencyRatio of realised multiple to peak multiple (0–1)
profit_factorTotal profit divided by total loss
return_stddevStandard deviation of per-trade returns
avg_loss_pct_on_losersAverage loss magnitude on losing trades
avg_gain_pct_on_winnersAverage gain magnitude on winning trades
is_bot1 if the wallet has been classified as a bot by BotDetector

Wallet context (1 feature)

FeatureDescription
wallet_age_daysDays since the wallet was first observed by the pipeline

Signal-level (10 features)

These features describe the specific buy event that triggered the signal, not the wallet’s history. They capture what is happening at this token, at this moment, for this particular buy.
FeatureDescription
buy_rankThis wallet’s buy rank on this token
sol_amountSize of this specific buy in SOL
curve_pct_at_buyBonding curve fill percentage at time of buy (0–1)
curve_solCurve SOL reserves at time of buy
velocity_buys_60sNumber of buys on this token in the last 60 seconds
velocity_buys_300sNumber of buys on this token in the last 300 seconds
sol_volume_60sSOL volume on this token in the last 60 seconds
buy_rank_percentileThis wallet’s buy rank divided by unique buyer count (0–1)
sol_vs_wallet_avgThis buy’s SOL amount relative to the wallet’s average (1.0 = typical size)
token_age_at_signalToken age in seconds at time of signal

Temporal (3 features)

These features encode time-of-day and day-of-week patterns. Market dynamics on Pump.fun vary significantly by time: weekend afternoons have different volume profiles than weekday mornings.
The source documentation labels this category as containing 2 features, but three are defined in the feature order.
FeatureDescription
buy_sell_ratioTotal buys divided by (total sells + 1) at time of signal
hour_of_dayUTC hour of the signal (0–23)
day_of_weekDay of week (0 = Sunday, 6 = Saturday)

Creator intelligence (8 features)

Sourced from CreatorRiskScorer, which tracks historical performance and behaviour patterns for every token creator the pipeline has observed.
FeatureDescription
creator_graduation_rateThis token creator’s historical graduation rate across all their tokens
creator_rug_rateQuick-death rate across this creator’s tokens
creator_risk_score0–100 composite creator risk score
creator_tokens_createdTotal tokens created by this creator
creator_serial_velocityTokens created per day over the last 30 days
creator_avg_insider_pctAverage insider presence across the creator’s tokens
creator_avg_bot_pctAverage bot buyer percentage across the creator’s tokens
creator_is_serial1 if the creator has been classified as a serial launcher

Token state (7 features)

These features capture the aggregate state of the token at the moment the signal fires, describing how much activity has accumulated and what risk indicators are present.
FeatureDescription
token_unique_buyersUnique buyers on this token so far
token_total_buysTotal buy transactions on this token
token_total_sellsTotal sell transactions on this token
token_risk_scoreToken-level risk score (0–100) from RiskScorer
token_bot_buyer_pctFraction of buyers on this token classified as bots
token_top10_concentrationShare of token supply held by the top 10 wallets
token_bundle_confidenceHighest bundle confidence score detected for this token

Lifecycle (1 feature)

FeatureDescription
lifecycle_state_encodedToken lifecycle state encoded as an integer (0 = launch … 7 = graduated)

Social graph (5 features)

Sourced from the CoOccurrence and GraphBuilder services, which track which wallets tend to buy the same tokens together and build a cluster graph from that data.
The source documentation labels this category as containing 4 features, but five are defined in the feature order.
FeatureDescription
cluster_sizeNumber of wallets in this wallet’s co-occurrence cluster
co_occurrence_max_scoreHighest co-occurrence edge score for this wallet
cluster_avg_grad_rateAverage graduation rate of wallets in the same cluster
tracked_wallets_already_inCount of tracked wallets already holding this token when the signal fires
is_first_tracked_buy1 if this is the first tracked wallet to buy this token

Market context (2 features)

These features encode the broader market environment at signal time. A signal that looks identical in isolation may have very different expected outcomes depending on whether the market is in a hot creation period or a slow one.
FeatureDescription
tokens_created_last_hourMarket-wide token creation rate over the last hour
rolling_graduation_rate_2hMarket-wide graduation rate over the last 2 hours

Default values

Every feature has a carefully chosen default value applied when the underlying data is unavailable — for example, when a signal fires from a wallet that has never been seen before, or when a token is too new to have velocity data.
Default values represent a neutral, unknown wallet in a neutral market — not a worst-case assumption. Using worst-case defaults would introduce systematic pessimism bias: the model would learn to treat unknown signals as bad signals, which is not correct. An unseen wallet could be excellent.
Defaults are chosen to be plausible midpoints within each feature’s observed distribution. For example:
  • wallet_avg_buy_rank defaults to 50 (median rank)
  • avg_sol_per_buy defaults to 0.5 SOL (typical small buy)
  • lifecycle_state_encoded defaults to -1 (unknown / not yet classified)
The models were trained with these same defaults applied whenever data was missing during training. This means the model has learned to handle the default values correctly and will not produce anomalous outputs when they appear at inference time.

Model architecture

How the feature vector is assembled into an ONNX input tensor and passed through Platt calibration.

Training methodology

How the 68-feature dataset is constructed, labelled, and used to train each model.

Build docs developers (and LLMs) love